Merge remote-tracking branch 'lnis_origin/dev' into ganesh_dev

openfpga/CMakeLists.txt

@@ -22,6 +22,7 @@ target_link_libraries(libopenfpga
                       libarchopenfpga
                       libopenfpgashell
                       libopenfpgautil
+                      libini
                       libvtrutil
                       libvpr8)
 

openfpga/src/base/io_location_map.cpp

@@ -0,0 +1,46 @@
+/******************************************************************************
+ * Memember functions for data structure IoLocationMap
+ ******************************************************************************/
+#include "vtr_assert.h"
+
+#include "io_location_map.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/**************************************************
+ * Public Accessors 
+ *************************************************/
+size_t IoLocationMap::io_index(const size_t& x, const size_t& y, const size_t& z) const {
+  if (x >= io_indices_.size()) {
+    return size_t(-1);
+  }
+
+  if (y >= io_indices_[x].size()) {
+    return size_t(-1);
+  }
+
+  if (z >= io_indices_[x][y].size()) {
+    return size_t(-1);
+  }
+
+  return io_indices_[x][y][z];
+}
+
+void IoLocationMap::set_io_index(const size_t& x, const size_t& y, const size_t& z, const size_t& io_index) {
+  if (x >= io_indices_.size()) {
+    io_indices_.resize(x + 1);
+  }
+
+  if (y >= io_indices_[x].size()) {
+    io_indices_[x].resize(y + 1);
+  }
+
+  if (z >= io_indices_[x][y].size()) {
+    io_indices_[x][y].resize(z + 1);
+  }
+
+  io_indices_[x][y][z] = io_index;
+}
+
+} /* end namespace openfpga */

openfpga/src/base/io_location_map.h

@@ -0,0 +1,39 @@
+#ifndef IO_LOCATION_MAP_H
+#define IO_LOCATION_MAP_H
+
+/********************************************************************
+ * Include header files required by the data structure definition
+ *******************************************************************/
+#include <vector>
+
+/* Begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * I/O location map is a data structure to bridge the index of I/Os
+ * in the FPGA fabric, i.e., the module graph, and logical location
+ * of the I/O in VPR coordinate system
+ *
+ * For example 
+ *              io[0]   io[1]          io[2]
+ *           +-----------------+    +--------+
+ *           |        |        |    |        |
+ *           |  I/O   |  I/O   |    |  I/O   |
+ *           | [0][y] | [0][y] |    | [1][y] |
+ *           |  [0]   |  [1]   |    |  [0]   |
+ *           +-----------------+    +--------+
+ *
+ *******************************************************************/
+class IoLocationMap {
+  public: /* Public aggregators */
+    size_t io_index(const size_t& x, const size_t& y, const size_t& z) const;
+  public: /* Public mutators */
+    void set_io_index(const size_t& x, const size_t& y, const size_t& z, const size_t& io_index);
+  private: /* Internal Data */
+    /* I/O index fast lookup by [x][y][z] location */
+    std::vector<std::vector<std::vector<size_t>>> io_indices_;
+};
+
+} /* End namespace openfpga*/
+
+#endif

openfpga/src/base/openfpga_bitstream.cpp

@@ -5,6 +5,9 @@
 #include "vtr_time.h"
 #include "vtr_log.h"
 
+/* Headers from openfpgautil library */
+#include "openfpga_digest.h"
+
 #include "build_device_bitstream.h"
 #include "bitstream_writer.h"
 #include "build_fabric_bitstream.h"
@@ -30,6 +33,11 @@ void fpga_bitstream(OpenfpgaContext& openfpga_ctx,
                                                                     cmd_context.option_enable(cmd, opt_verbose));
 
   if (true == cmd_context.option_enable(cmd, opt_file)) {
+    std::string src_dir_path = find_path_dir_name(cmd_context.option_value(cmd, opt_file));
+
+    /* Create directories */
+    create_dir_path(src_dir_path.c_str());
+
     write_arch_independent_bitstream_to_xml_file(openfpga_ctx.bitstream_manager(),
                                                  cmd_context.option_value(cmd, opt_file));
   }

openfpga/src/base/openfpga_build_fabric.cpp

@@ -21,43 +21,43 @@ namespace openfpga {
  * This function should only be called after the GSB builder is done
  *******************************************************************/
 static 
-void compress_routing_hierarchy(OpenfpgaContext& openfpga_context,
+void compress_routing_hierarchy(OpenfpgaContext& openfpga_ctx,
                                 const bool& verbose_output) {
   vtr::ScopedStartFinishTimer timer("Identify unique General Switch Blocks (GSBs)");
 
   /* Build unique module lists */
-  openfpga_context.mutable_device_rr_gsb().build_unique_module(g_vpr_ctx.device().rr_graph);
+  openfpga_ctx.mutable_device_rr_gsb().build_unique_module(g_vpr_ctx.device().rr_graph);
 
   /* Report the stats */
   VTR_LOGV(verbose_output, 
            "Detected %lu unique X-direction connection blocks from a total of %d (compression rate=%d%)\n",
-           openfpga_context.device_rr_gsb().get_num_cb_unique_module(CHANX),
+           openfpga_ctx.device_rr_gsb().get_num_cb_unique_module(CHANX),
-           find_device_rr_gsb_num_cb_modules(openfpga_context.device_rr_gsb(), CHANX),
+           find_device_rr_gsb_num_cb_modules(openfpga_ctx.device_rr_gsb(), CHANX),
-           100 * (openfpga_context.device_rr_gsb().get_num_cb_unique_module(CHANX) / find_device_rr_gsb_num_cb_modules(openfpga_context.device_rr_gsb(), CHANX) - 1));
+           100 * (openfpga_ctx.device_rr_gsb().get_num_cb_unique_module(CHANX) / find_device_rr_gsb_num_cb_modules(openfpga_ctx.device_rr_gsb(), CHANX) - 1));
 
   VTR_LOGV(verbose_output,
            "Detected %lu unique Y-direction connection blocks from a total of %d (compression rate=%d%)\n",
-           openfpga_context.device_rr_gsb().get_num_cb_unique_module(CHANY),
+           openfpga_ctx.device_rr_gsb().get_num_cb_unique_module(CHANY),
-           find_device_rr_gsb_num_cb_modules(openfpga_context.device_rr_gsb(), CHANY),
+           find_device_rr_gsb_num_cb_modules(openfpga_ctx.device_rr_gsb(), CHANY),
-           100 * (openfpga_context.device_rr_gsb().get_num_cb_unique_module(CHANY) / find_device_rr_gsb_num_cb_modules(openfpga_context.device_rr_gsb(), CHANY) - 1));
+           100 * (openfpga_ctx.device_rr_gsb().get_num_cb_unique_module(CHANY) / find_device_rr_gsb_num_cb_modules(openfpga_ctx.device_rr_gsb(), CHANY) - 1));
 
   VTR_LOGV(verbose_output,
            "Detected %lu unique switch blocks from a total of %d (compression rate=%d%)\n",
-           openfpga_context.device_rr_gsb().get_num_sb_unique_module(),
+           openfpga_ctx.device_rr_gsb().get_num_sb_unique_module(),
-           find_device_rr_gsb_num_sb_modules(openfpga_context.device_rr_gsb()),
+           find_device_rr_gsb_num_sb_modules(openfpga_ctx.device_rr_gsb()),
-           100 * (openfpga_context.device_rr_gsb().get_num_sb_unique_module() / find_device_rr_gsb_num_sb_modules(openfpga_context.device_rr_gsb()) - 1));
+           100 * (openfpga_ctx.device_rr_gsb().get_num_sb_unique_module() / find_device_rr_gsb_num_sb_modules(openfpga_ctx.device_rr_gsb()) - 1));
 
   VTR_LOGV(verbose_output,
            "Detected %lu unique general switch blocks from a total of %d (compression rate=%d%)\n",
-           openfpga_context.device_rr_gsb().get_num_gsb_unique_module(),
+           openfpga_ctx.device_rr_gsb().get_num_gsb_unique_module(),
-           find_device_rr_gsb_num_gsb_modules(openfpga_context.device_rr_gsb()),
+           find_device_rr_gsb_num_gsb_modules(openfpga_ctx.device_rr_gsb()),
-           100 * (openfpga_context.device_rr_gsb().get_num_gsb_unique_module() / find_device_rr_gsb_num_gsb_modules(openfpga_context.device_rr_gsb()) - 1));
+           100 * (openfpga_ctx.device_rr_gsb().get_num_gsb_unique_module() / find_device_rr_gsb_num_gsb_modules(openfpga_ctx.device_rr_gsb()) - 1));
 }
 
 /********************************************************************
  * Build the module graph for FPGA device
  *******************************************************************/
-void build_fabric(OpenfpgaContext& openfpga_context,
+void build_fabric(OpenfpgaContext& openfpga_ctx,
                   const Command& cmd, const CommandContext& cmd_context) { 
 
   CommandOptionId opt_compress_routing = cmd.option("compress_routing");
@@ -65,13 +65,14 @@ void build_fabric(OpenfpgaContext& openfpga_context,
   CommandOptionId opt_verbose = cmd.option("verbose");
   
   if (true == cmd_context.option_enable(cmd, opt_compress_routing)) {
-    compress_routing_hierarchy(openfpga_context, cmd_context.option_enable(cmd, opt_verbose));
+    compress_routing_hierarchy(openfpga_ctx, cmd_context.option_enable(cmd, opt_verbose));
   }
 
   VTR_LOG("\n");
 
-  openfpga_context.mutable_module_graph() = build_device_module_graph(g_vpr_ctx.device(),
+  openfpga_ctx.mutable_module_graph() = build_device_module_graph(openfpga_ctx.mutable_io_location_map(),
-                                                                      const_cast<const OpenfpgaContext&>(openfpga_context),
+                                                                  g_vpr_ctx.device(),
+                                                                  const_cast<const OpenfpgaContext&>(openfpga_ctx),
                                                                   cmd_context.option_enable(cmd, opt_compress_routing),
                                                                   cmd_context.option_enable(cmd, opt_duplicate_grid_pin),
                                                                   cmd_context.option_enable(cmd, opt_verbose));

openfpga/src/base/openfpga_build_fabric.h

@@ -15,7 +15,7 @@
 /* begin namespace openfpga */
 namespace openfpga {
 
-void build_fabric(OpenfpgaContext& openfpga_context,
+void build_fabric(OpenfpgaContext& openfpga_ctx,
                   const Command& cmd, const CommandContext& cmd_context); 
 
 } /* end namespace openfpga */

openfpga/src/base/openfpga_context.h

@@ -15,6 +15,7 @@
 #include "openfpga_flow_manager.h"
 #include "bitstream_manager.h"
 #include "device_rr_gsb.h"
+#include "io_location_map.h"
 
 /********************************************************************
  * This file includes the declaration of the date structure 
@@ -58,6 +59,8 @@ class OpenfpgaContext : public Context  {
     const openfpga::FlowManager& flow_manager() const { return flow_manager_; }
     const openfpga::BitstreamManager& bitstream_manager() const { return bitstream_manager_; }
     const std::vector<openfpga::ConfigBitId>& fabric_bitstream() const { return fabric_bitstream_; }
+    const openfpga::IoLocationMap& io_location_map() { return io_location_map_; }
+    const std::unordered_map<AtomNetId, t_net_power>& net_activity() { return net_activity_; }
   public:  /* Public mutators */
     openfpga::Arch& mutable_arch() { return arch_; }
     openfpga::VprDeviceAnnotation& mutable_vpr_device_annotation() { return vpr_device_annotation_; }
@@ -72,6 +75,8 @@ class OpenfpgaContext : public Context  {
     openfpga::FlowManager& mutable_flow_manager() { return flow_manager_; }
     openfpga::BitstreamManager& mutable_bitstream_manager() { return bitstream_manager_; }
     std::vector<openfpga::ConfigBitId>& mutable_fabric_bitstream() { return fabric_bitstream_; }
+    openfpga::IoLocationMap& mutable_io_location_map() { return io_location_map_; }
+    std::unordered_map<AtomNetId, t_net_power>& mutable_net_activity() { return net_activity_; }
   private: /* Internal data */
     /* Data structure to store information from read_openfpga_arch library */
     openfpga::Arch arch_;
@@ -102,11 +107,15 @@ class OpenfpgaContext : public Context  {
 
     /* Fabric module graph */
     openfpga::ModuleManager module_graph_;
+    openfpga::IoLocationMap io_location_map_;
 
     /* Bitstream database */
     openfpga::BitstreamManager bitstream_manager_;
     std::vector<openfpga::ConfigBitId> fabric_bitstream_;
 
+    /* Net activities of users' implementation */
+    std::unordered_map<AtomNetId, t_net_power> net_activity_; 
+ 
     /* Flow status */
     openfpga::FlowManager flow_manager_;
 };

openfpga/src/base/openfpga_link_arch.cpp

@@ -2,11 +2,20 @@
  * This file includes functions to read an OpenFPGA architecture file
  * which are built on the libarchopenfpga library
  *******************************************************************/
+#include <cmath>
+#include <iterator>
+
 /* Headers from vtrutil library */
 #include "vtr_time.h"
 #include "vtr_assert.h"
 #include "vtr_log.h"
 
+/* Headers from vpr library */
+#include "timing_info.h"
+#include "AnalysisDelayCalculator.h"
+#include "net_delay.h"
+#include "read_activity.h"
+
 #include "vpr_device_annotation.h"
 #include "pb_type_utils.h"
 #include "annotate_pb_types.h"
@@ -46,6 +55,163 @@ bool is_vpr_rr_graph_supported(const RRGraph& rr_graph) {
   return true;
 }
 
+/********************************************************************
+ * Find the number of clock cycles in simulation based on the average signal density
+ *******************************************************************/
+static 
+size_t recommend_num_sim_clock_cycle(const AtomContext& atom_ctx,
+                                     const std::unordered_map<AtomNetId, t_net_power>& net_activity, 
+                                     const float& sim_window_size) {
+  size_t recmd_num_sim_clock_cycle = 0;
+
+  float avg_density = 0.;
+  size_t net_cnt = 0;
+
+  float weighted_avg_density = 0.;
+  size_t weighted_net_cnt = 0;
+
+  /* get the average density of all the nets */
+  for (const AtomNetId& atom_net : atom_ctx.nlist.nets()) {
+    /* Skip the nets without any activity annotation */
+    if (0 == net_activity.count(atom_net)) {
+      continue;
+    }
+
+    /* Only care non-zero density nets */
+    if (0. == net_activity.at(atom_net).density) {
+      continue;
+    }
+
+    avg_density += net_activity.at(atom_net).density; 
+    net_cnt++;
+
+    /* Consider the weight of fan-out */
+    size_t net_weight; 
+    if (0 == std::distance(atom_ctx.nlist.net_sinks(atom_net).begin(), atom_ctx.nlist.net_sinks(atom_net).end())) {
+      net_weight = 1;
+    } else {
+      VTR_ASSERT(0 < std::distance(atom_ctx.nlist.net_sinks(atom_net).begin(), atom_ctx.nlist.net_sinks(atom_net).end()));
+      net_weight = std::distance(atom_ctx.nlist.net_sinks(atom_net).begin(), atom_ctx.nlist.net_sinks(atom_net).end());
+    }
+    weighted_avg_density += net_activity.at(atom_net).density* net_weight;
+    weighted_net_cnt += net_weight;
+  }
+  avg_density = avg_density / net_cnt; 
+  weighted_avg_density = weighted_avg_density / weighted_net_cnt; 
+
+  /* Sort the net density */
+  std::vector<float> net_densities;
+  net_densities.reserve(net_cnt);
+  for (const AtomNetId& atom_net : atom_ctx.nlist.nets()) {
+    /* Skip the nets without any activity annotation */
+    if (0 == net_activity.count(atom_net)) {
+      continue;
+    }
+
+    /* Only care non-zero density nets */
+    if (0. == net_activity.at(atom_net).density) {
+      continue;
+    }
+
+    net_densities.push_back(net_activity.at(atom_net).density);
+  }
+  std::sort(net_densities.begin(), net_densities.end());
+  /* Get the median */
+  float median_density = 0.;
+  /* check for even case */
+  if (net_cnt % 2 != 0) { 
+    median_density = net_densities[size_t(net_cnt / 2)];
+  } else {              
+    median_density = 0.5 * (net_densities[size_t((net_cnt - 1) / 2)] + net_densities[size_t((net_cnt - 1) / 2)]);
+  }
+
+  /* It may be more reasonable to use median 
+   * But, if median density is 0, we use average density
+  */
+  if ((0. == median_density) && (0. == avg_density)) {
+    recmd_num_sim_clock_cycle = 1;
+    VTR_LOG_WARN("All the signal density is zero!\nNumber of clock cycles in simulations are set to be %ld!\n",
+                 recmd_num_sim_clock_cycle);
+  } else if (0. == avg_density) {
+      recmd_num_sim_clock_cycle = (int)round(1 / median_density); 
+  } else if (0. == median_density) {
+      recmd_num_sim_clock_cycle = (int)round(1 / avg_density);
+  } else {
+    /* add a sim window size to balance the weight of average density and median density
+     * In practice, we find that there could be huge difference between avereage and median values 
+     * For a reasonable number of simulation clock cycles, we do this window size.
+     */
+    recmd_num_sim_clock_cycle = (int)round(1 / (sim_window_size * avg_density + (1 - sim_window_size) * median_density ));
+  }
+  
+  VTR_ASSERT(0 < recmd_num_sim_clock_cycle);
+
+  VTR_LOG("Average net density: %.2f\n", avg_density);
+  VTR_LOG("Median net density: %.2f\n", median_density);
+  VTR_LOG("Average net density after weighting: %.2f\n", weighted_avg_density);
+  VTR_LOG("Window size set for Simulation: %.2f\n", sim_window_size);
+  VTR_LOG("Net density after Window size : %.2f\n", 
+          (sim_window_size * avg_density + (1 - sim_window_size) * median_density));
+  VTR_LOG("Recommend no. of clock cycles: %ld\n", recmd_num_sim_clock_cycle);
+
+  return recmd_num_sim_clock_cycle; 
+}
+
+/********************************************************************
+ * Annotate simulation setting based on VPR results
+ *  - If the operating clock frequency is set to follow the vpr timing results,
+ *    we will set a new operating clock frequency here
+ *  - If the number of clock cycles in simulation is set to be automatically determined,
+ *    we will infer the number based on the average signal density
+ *******************************************************************/
+static 
+void annotate_simulation_setting(const AtomContext& atom_ctx, 
+                                 const std::unordered_map<AtomNetId, t_net_power>& net_activity, 
+                                 SimulationSetting& sim_setting) {
+
+  /* Find if the operating frequency is binded to vpr results */
+  if (0. == sim_setting.operating_clock_frequency()) {
+    VTR_LOG("User specified the operating clock frequency to use VPR results\n");
+    /* Run timing analysis and collect critical path delay
+     * This code is copied from function vpr_analysis() in vpr_api.h 
+     * Should keep updated to latest VPR code base
+     * Note:
+     *   - MUST mention in documentation that VPR should be run in timing enabled mode
+     */
+    vtr::vector<ClusterNetId, float*> net_delay;
+    vtr::t_chunk net_delay_ch;
+    /* Load the net delays */
+    net_delay = alloc_net_delay(&net_delay_ch);
+    load_net_delay_from_routing(net_delay);
+
+    /* Do final timing analysis */
+    auto analysis_delay_calc = std::make_shared<AnalysisDelayCalculator>(atom_ctx.nlist, atom_ctx.lookup, net_delay);
+    auto timing_info = make_setup_hold_timing_info(analysis_delay_calc);
+    timing_info->update();
+
+    /* Get critical path delay. Update simulation settings */
+    float T_crit = timing_info->least_slack_critical_path().delay() * (1. + sim_setting.operating_clock_frequency_slack());
+    sim_setting.set_operating_clock_frequency(1 / T_crit); 
+    VTR_LOG("Use VPR critical path delay %g [ns] with a %g [%] slack in OpenFPGA.\n",
+            T_crit / 1e9, sim_setting.operating_clock_frequency_slack() * 100);
+  }
+  VTR_LOG("Will apply operating clock frequency %g [MHz] to simulations\n",
+          sim_setting.operating_clock_frequency() / 1e6);
+
+  if (0. == sim_setting.num_clock_cycles()) {
+    /* Find the number of clock cycles to be used in simulation by average over the signal activity */
+
+    VTR_LOG("User specified the number of operating clock cycles to be inferred from signal activities\n");
+    size_t num_clock_cycles = recommend_num_sim_clock_cycle(atom_ctx,
+                                                            net_activity, 
+                                                            0.5);
+    sim_setting.set_num_clock_cycles(num_clock_cycles);
+
+    VTR_LOG("Will apply %lu operating clock cycles to simulations\n",
+            sim_setting.num_clock_cycles());
+  }
+}
+
 /********************************************************************
  * Top-level function to link openfpga architecture to VPR, including:
  * - physical pb_type
@@ -59,6 +225,7 @@ void link_arch(OpenfpgaContext& openfpga_ctx,
 
   vtr::ScopedStartFinishTimer timer("Link OpenFPGA architecture to VPR architecture");
 
+  CommandOptionId opt_activity_file = cmd.option("activity_file");
   CommandOptionId opt_verbose = cmd.option("verbose");
 
   /* Annotate pb_type graphs
@@ -118,6 +285,22 @@ void link_arch(OpenfpgaContext& openfpga_ctx,
                          g_vpr_ctx.clustering(),
                          g_vpr_ctx.placement(),
                          openfpga_ctx.mutable_vpr_placement_annotation());
+
+  /* Read activity file is manadatory in the following flow-run settings
+   * - When users specify that number of clock cycles 
+   *   should be inferred from FPGA implmentation
+   * - When FPGA-SPICE is enabled
+   */
+  openfpga_ctx.mutable_net_activity() = read_activity(g_vpr_ctx.atom().nlist,
+                                                      cmd_context.option_value(cmd, opt_activity_file).c_str());
+
+  /* TODO: Annotate the number of clock cycles and clock frequency by following VPR results
+   * We SHOULD create a new simulation setting for OpenFPGA use only
+   * Avoid overwrite the raw data achieved when parsing!!!
+   */
+  annotate_simulation_setting(g_vpr_ctx.atom(),
+                              openfpga_ctx.net_activity(),
+                              openfpga_ctx.mutable_arch().sim_setting);
 } 
 
 } /* end namespace openfpga */

openfpga/src/base/openfpga_sdc.cpp

@@ -0,0 +1,80 @@
+/********************************************************************
+ * This file includes functions to compress the hierachy of routing architecture
+ *******************************************************************/
+/* Headers from vtrutil library */
+#include "vtr_time.h"
+#include "vtr_log.h"
+
+/* Headers from openfpgautil library */
+#include "openfpga_digest.h"
+
+#include "circuit_library_utils.h"
+#include "pnr_sdc_writer.h"
+#include "openfpga_sdc.h"
+
+/* Include global variables of VPR */
+#include "globals.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * A wrapper function to call the PnR SDC generator of FPGA-SDC
+ *******************************************************************/
+void write_pnr_sdc(OpenfpgaContext& openfpga_ctx,
+                   const Command& cmd, const CommandContext& cmd_context) {
+
+  CommandOptionId opt_output_dir = cmd.option("file");
+  CommandOptionId opt_constrain_global_port = cmd.option("constrain_global_port");
+  CommandOptionId opt_constrain_grid = cmd.option("constrain_grid");
+  CommandOptionId opt_constrain_sb = cmd.option("constrain_sb");
+  CommandOptionId opt_constrain_cb = cmd.option("constrain_cb");
+  CommandOptionId opt_constrain_configurable_memory_outputs = cmd.option("constrain_configurable_memory_outputs");
+  CommandOptionId opt_constrain_routing_multiplexer_outputs = cmd.option("constrain_routing_multiplexer_outputs");
+  CommandOptionId opt_constrain_switch_block_outputs = cmd.option("constrain_switch_block_outputs");
+
+  /* This is an intermediate data structure which is designed to modularize the FPGA-SDC
+   * Keep it independent from any other outside data structures
+   */
+  std::string sdc_dir_path = format_dir_path(cmd_context.option_value(cmd, opt_output_dir));
+
+  /* Create directories */
+  create_dir_path(sdc_dir_path.c_str());
+
+  PnrSdcOption options(sdc_dir_path);
+
+  options.set_constrain_global_port(cmd_context.option_enable(cmd, opt_constrain_global_port));
+  options.set_constrain_grid(cmd_context.option_enable(cmd, opt_constrain_grid));
+  options.set_constrain_sb(cmd_context.option_enable(cmd, opt_constrain_sb));
+  options.set_constrain_cb(cmd_context.option_enable(cmd, opt_constrain_cb));
+  options.set_constrain_configurable_memory_outputs(cmd_context.option_enable(cmd, opt_constrain_configurable_memory_outputs));
+  options.set_constrain_routing_multiplexer_outputs(cmd_context.option_enable(cmd, opt_constrain_routing_multiplexer_outputs));
+  options.set_constrain_switch_block_outputs(cmd_context.option_enable(cmd, opt_constrain_switch_block_outputs));
+
+  /* We first turn on default sdc option and then disable part of them by following users' options */
+  if (false == options.generate_sdc_pnr()) {
+    options.set_generate_sdc_pnr(true);
+  }
+
+  /* Collect global ports from the circuit library:
+   * TODO: should we place this in the OpenFPGA context?
+   */
+  std::vector<CircuitPortId> global_ports = find_circuit_library_global_ports(openfpga_ctx.arch().circuit_lib);
+  
+  /* Execute only when sdc is enabled */
+  if (true == options.generate_sdc_pnr()) { 
+    print_pnr_sdc(options,
+                  1./openfpga_ctx.arch().sim_setting.programming_clock_frequency(),
+                  1./openfpga_ctx.arch().sim_setting.operating_clock_frequency(),
+                  g_vpr_ctx.device(),
+                  openfpga_ctx.vpr_device_annotation(),
+                  openfpga_ctx.device_rr_gsb(),
+                  openfpga_ctx.module_graph(),
+                  openfpga_ctx.mux_lib(),
+                  openfpga_ctx.arch().circuit_lib,
+                  global_ports,
+                  openfpga_ctx.flow_manager().compress_routing());
+  }
+} 
+
+} /* end namespace openfpga */

openfpga/src/base/openfpga_sdc.h

@@ -0,0 +1,23 @@
+#ifndef OPENFPGA_SDC_H
+#define OPENFPGA_SDC_H
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+#include "command.h"
+#include "command_context.h"
+#include "openfpga_context.h"
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+void write_pnr_sdc(OpenfpgaContext& openfpga_ctx,
+                   const Command& cmd, const CommandContext& cmd_context); 
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/base/openfpga_sdc_command.cpp

@@ -0,0 +1,79 @@
+/********************************************************************
+ * Add commands to the OpenFPGA shell interface, 
+ * in purpose of generate SDC files
+ * - write_pnr_sdc : generate SDC to constrain the back-end flow for FPGA fabric
+ * - write_analysis_sdc: TODO: generate SDC based on users' implementations
+ *******************************************************************/
+#include "openfpga_sdc.h"
+#include "openfpga_sdc_command.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * - Add a command to Shell environment: generate PnR SDC 
+ * - Add associated options 
+ * - Add command dependency
+ *******************************************************************/
+static 
+void add_openfpga_write_pnr_sdc_command(openfpga::Shell<OpenfpgaContext>& shell,
+                                        const ShellCommandClassId& cmd_class_id,
+                                        const ShellCommandId& shell_cmd_build_fabric_id) {
+  Command shell_cmd("write_pnr_sdc");
+
+  /* Add an option '--file' in short '-f'*/
+  CommandOptionId output_opt = shell_cmd.add_option("file", true, "Specify the output directory for SDC files");
+  shell_cmd.set_option_short_name(output_opt, "f");
+  shell_cmd.set_option_require_value(output_opt, openfpga::OPT_STRING);
+
+  /* Add an option '--constrain_global_port' */
+  shell_cmd.add_option("constrain_global_port", false, "Constrain all the global ports of FPGA fabric");
+
+  /* Add an option '--constrain_grid' */
+  shell_cmd.add_option("constrain_grid", false, "Constrain all the grids of FPGA fabric");
+
+  /* Add an option '--constrain_sb' */
+  shell_cmd.add_option("constrain_sb", false, "Constrain all the switch blocks of FPGA fabric");
+
+  /* Add an option '--constrain_cb' */
+  shell_cmd.add_option("constrain_cb", false, "Constrain all the connection blocks of FPGA fabric");
+
+  /* Add an option '--constrain_configurable_memory_outputs' */
+  shell_cmd.add_option("constrain_configurable_memory_outputs", false, "Constrain all the outputs of configurable memories of FPGA fabric");
+
+  /* Add an option '--constrain_routing_multiplexer_outputs' */
+  shell_cmd.add_option("constrain_routing_multiplexer_outputs", false, "Constrain all the outputs of routing multiplexer of FPGA fabric");
+
+  /* Add an option '--constrain_switch_block_outputs' */
+  shell_cmd.add_option("constrain_switch_block_outputs", false, "Constrain all the outputs of switch blocks of FPGA fabric");
+
+  /* Add an option '--verbose' */
+  shell_cmd.add_option("verbose", false, "Enable verbose output");
+  
+  /* Add command 'write_fabric_verilog' to the Shell */
+  ShellCommandId shell_cmd_id = shell.add_command(shell_cmd, "generate SDC files to constrain the backend flow for FPGA fabric");
+  shell.set_command_class(shell_cmd_id, cmd_class_id);
+  shell.set_command_execute_function(shell_cmd_id, write_pnr_sdc);
+
+  /* The 'build_fabric' command should NOT be executed before 'link_openfpga_arch' */
+  std::vector<ShellCommandId> cmd_dependency;
+  cmd_dependency.push_back(shell_cmd_build_fabric_id);
+  shell.set_command_dependency(shell_cmd_id, cmd_dependency);
+}
+
+void add_openfpga_sdc_commands(openfpga::Shell<OpenfpgaContext>& shell) {
+  /* Get the unique id of 'build_fabric' command which is to be used in creating the dependency graph */
+  const ShellCommandId& shell_cmd_build_fabric_id = shell.command(std::string("build_fabric"));
+
+  /* Add a new class of commands */
+  ShellCommandClassId openfpga_sdc_cmd_class = shell.add_command_class("FPGA-SDC");
+
+  /******************************** 
+   * Command 'write_fabric_verilog' 
+   */
+  add_openfpga_write_pnr_sdc_command(shell,
+                                     openfpga_sdc_cmd_class,
+                                     shell_cmd_build_fabric_id);
+} 
+
+} /* end namespace openfpga */

openfpga/src/base/openfpga_sdc_command.h

@@ -0,0 +1,21 @@
+#ifndef OPENFPGA_SDC_COMMAND_H
+#define OPENFPGA_SDC_COMMAND_H
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+#include "shell.h"
+#include "openfpga_context.h"
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+void add_openfpga_sdc_commands(openfpga::Shell<OpenfpgaContext>& shell); 
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/base/openfpga_setup_command.cpp

@@ -14,9 +14,199 @@
 /* begin namespace openfpga */
 namespace openfpga {
 
+/********************************************************************
+ * - Add a command to Shell environment: read_openfpga_arch
+ * - Add associated options 
+ * - Add command dependency
+ *******************************************************************/
+static 
+ShellCommandId add_openfpga_read_arch_command(openfpga::Shell<OpenfpgaContext>& shell,
+                                              const ShellCommandClassId& cmd_class_id) {
+  Command shell_cmd("read_openfpga_arch");
+
+  /* Add an option '--file' in short '-f'*/
+  CommandOptionId opt_arch_file = shell_cmd.add_option("file", true, "file path to the architecture XML");
+  shell_cmd.set_option_short_name(opt_arch_file, "f");
+  shell_cmd.set_option_require_value(opt_arch_file, openfpga::OPT_STRING);
+
+  /* Add command 'read_openfpga_arch' to the Shell */
+  ShellCommandId shell_cmd_id = shell.add_command(shell_cmd, "read OpenFPGA architecture file");
+  shell.set_command_class(shell_cmd_id, cmd_class_id);
+  shell.set_command_execute_function(shell_cmd_id, read_arch);
+
+  return shell_cmd_id;
+}
+
+/********************************************************************
+ * - Add a command to Shell environment: write_openfpga_arch
+ * - Add associated options 
+ * - Add command dependency
+ *******************************************************************/
+static 
+ShellCommandId add_openfpga_write_arch_command(openfpga::Shell<OpenfpgaContext>& shell,
+                                               const ShellCommandClassId& cmd_class_id,
+                                               const std::vector<ShellCommandId>& dependent_cmds) {
+  Command shell_cmd("write_openfpga_arch");
+  /* Add an option '--file' in short '-f'*/
+  CommandOptionId opt_file = shell_cmd.add_option("file", true, "file path to the architecture XML");
+  shell_cmd.set_option_short_name(opt_file, "f");
+  shell_cmd.set_option_require_value(opt_file, openfpga::OPT_STRING);
+
+  /* Add command 'write_openfpga_arch' to the Shell */
+  ShellCommandId shell_cmd_id = shell.add_command(shell_cmd, "write OpenFPGA architecture file");
+  shell.set_command_class(shell_cmd_id, cmd_class_id);
+  shell.set_command_const_execute_function(shell_cmd_id, write_arch);
+
+  /* The 'write_openfpga_arch' command should NOT be executed before 'read_openfpga_arch' */
+  shell.set_command_dependency(shell_cmd_id, dependent_cmds);
+
+  return shell_cmd_id;
+}
+
+/********************************************************************
+ * - Add a command to Shell environment: link_openfpga_arch
+ * - Add associated options 
+ * - Add command dependency
+ *******************************************************************/
+static 
+ShellCommandId add_openfpga_link_arch_command(openfpga::Shell<OpenfpgaContext>& shell,
+                                              const ShellCommandClassId& cmd_class_id,
+                                              const std::vector<ShellCommandId>& dependent_cmds) {
+  Command shell_cmd("link_openfpga_arch");
+
+  /* Add an option '--activity_file'*/
+  CommandOptionId opt_act_file = shell_cmd.add_option("activity_file", true, "file path to the signal activity");
+  shell_cmd.set_option_require_value(opt_act_file, openfpga::OPT_STRING);
+
+  /* Add an option '--verbose' */
+  shell_cmd.add_option("verbose", false, "Show verbose outputs");
+  
+  /* Add command 'link_openfpga_arch' to the Shell */
+  ShellCommandId shell_cmd_id = shell.add_command(shell_cmd, "Bind OpenFPGA architecture to VPR");
+  shell.set_command_class(shell_cmd_id, cmd_class_id);
+  shell.set_command_execute_function(shell_cmd_id, link_arch);
+
+  /* The 'link_openfpga_arch' command should NOT be executed before 'read_openfpga_arch' and 'vpr' */
+  shell.set_command_dependency(shell_cmd_id, dependent_cmds);
+
+  return shell_cmd_id;
+}
+
+/********************************************************************
+ * - Add a command to Shell environment: check_netlist_naming_conflict
+ * - Add associated options 
+ * - Add command dependency
+ *******************************************************************/
+static 
+ShellCommandId add_openfpga_check_netlist_naming_conflict_command(openfpga::Shell<OpenfpgaContext>& shell,
+                                                                  const ShellCommandClassId& cmd_class_id,
+                                                                  const std::vector<ShellCommandId>& dependent_cmds) {
+
+  Command shell_cmd("check_netlist_naming_conflict");
+
+  /* Add an option '--fix' */
+  shell_cmd.add_option("fix", false, "Apply correction to any conflicts found");
+
+  /* Add an option '--report' */
+  CommandOptionId opt_rpt = shell_cmd.add_option("report", false, "Output a report file about what any correction applied");
+  shell_cmd.set_option_require_value(opt_rpt, openfpga::OPT_STRING);
+
+  /* Add command 'check_netlist_naming_conflict' to the Shell */
+  ShellCommandId shell_cmd_id = shell.add_command(shell_cmd, "Check any block/net naming in users' BLIF netlist violates the syntax of fabric generator");
+  shell.set_command_class(shell_cmd_id, cmd_class_id);
+  shell.set_command_execute_function(shell_cmd_id, check_netlist_naming_conflict);
+
+  /* The 'link_openfpga_arch' command should NOT be executed before 'vpr' */
+  shell.set_command_dependency(shell_cmd_id, dependent_cmds);
+
+  return shell_cmd_id;
+}
+
+/********************************************************************
+ * - Add a command to Shell environment: pb_pin_fixup
+ * - Add associated options 
+ * - Add command dependency
+ *******************************************************************/
+static 
+ShellCommandId add_openfpga_pb_pin_fixup_command(openfpga::Shell<OpenfpgaContext>& shell,
+                                                 const ShellCommandClassId& cmd_class_id,
+                                                 const std::vector<ShellCommandId>& dependent_cmds) {
+
+  Command shell_cmd("pb_pin_fixup");
+  /* Add an option '--verbose' */
+  shell_cmd.add_option("verbose", false, "Show verbose outputs");
+
+  /* Add command 'pb_pin_fixup' to the Shell */
+  ShellCommandId shell_cmd_id = shell.add_command(shell_cmd, "Fix up the packing results due to pin swapping during routing stage");
+  shell.set_command_class(shell_cmd_id, cmd_class_id);
+  shell.set_command_execute_function(shell_cmd_id, pb_pin_fixup);
+
+  /* The 'pb_pin_fixup' command should NOT be executed before 'read_openfpga_arch' and 'vpr' */
+  shell.set_command_dependency(shell_cmd_id, dependent_cmds);
+
+  return shell_cmd_id;
+}
+
+/********************************************************************
+ * - Add a command to Shell environment: lut_truth_table_fixup
+ * - Add associated options 
+ * - Add command dependency
+ *******************************************************************/
+static 
+ShellCommandId add_openfpga_lut_truth_table_fixup_command(openfpga::Shell<OpenfpgaContext>& shell,
+                                                          const ShellCommandClassId& cmd_class_id,
+                                                          const std::vector<ShellCommandId>& dependent_cmds) {
+
+  Command shell_cmd("lut_truth_table_fixup");
+  /* Add an option '--verbose' */
+  shell_cmd.add_option("verbose", false, "Show verbose outputs");
+
+  /* Add command 'lut_truth_table_fixup' to the Shell */
+  ShellCommandId shell_cmd_id = shell.add_command(shell_cmd, "Fix up the truth table of Look-Up Tables due to pin swapping during packing stage");
+  shell.set_command_class(shell_cmd_id, cmd_class_id);
+  shell.set_command_execute_function(shell_cmd_id, lut_truth_table_fixup);
+
+  /* The 'lut_truth_table_fixup' command should NOT be executed before 'read_openfpga_arch' and 'vpr' */
+  shell.set_command_dependency(shell_cmd_id, dependent_cmds);
+
+  return shell_cmd_id;
+}
+
+/********************************************************************
+ * - Add a command to Shell environment: build_fabric
+ * - Add associated options 
+ * - Add command dependency
+ *******************************************************************/
+static 
+ShellCommandId add_openfpga_build_fabric_command(openfpga::Shell<OpenfpgaContext>& shell,
+                                                 const ShellCommandClassId& cmd_class_id,
+                                                 const std::vector<ShellCommandId>& dependent_cmds) {
+
+  Command shell_cmd("build_fabric");
+
+  /* Add an option '--compress_routing' */
+  shell_cmd.add_option("compress_routing", false, "Compress the number of unique routing modules by identifying the unique GSBs");
+
+  /* Add an option '--duplicate_grid_pin' */
+  shell_cmd.add_option("duplicate_grid_pin", false, "Duplicate the pins on the same side of a grid");
+
+  /* Add an option '--verbose' */
+  shell_cmd.add_option("verbose", false, "Show verbose outputs");
+
+  /* Add command 'compact_routing_hierarchy' to the Shell */
+  ShellCommandId shell_cmd_id = shell.add_command(shell_cmd, "Build the FPGA fabric in a graph of modules");
+  shell.set_command_class(shell_cmd_id, cmd_class_id);
+  shell.set_command_execute_function(shell_cmd_id, build_fabric);
+
+  /* The 'build_fabric' command should NOT be executed before 'link_openfpga_arch' */
+  shell.set_command_dependency(shell_cmd_id, dependent_cmds);
+
+  return shell_cmd_id;
+}
+
 void add_openfpga_setup_commands(openfpga::Shell<OpenfpgaContext>& shell) {
   /* Get the unique id of 'vpr' command which is to be used in creating the dependency graph */
-  const ShellCommandId& shell_cmd_vpr_id = shell.command(std::string("vpr"));
+  const ShellCommandId& vpr_cmd_id = shell.command(std::string("vpr"));
 
   /* Add a new class of commands */
   ShellCommandClassId openfpga_setup_cmd_class = shell.add_command_class("OpenFPGA setup");
@@ -24,119 +214,62 @@ void add_openfpga_setup_commands(openfpga::Shell<OpenfpgaContext>& shell) {
   /******************************** 
    * Command 'read_openfpga_arch' 
    */
-  Command shell_cmd_read_arch("read_openfpga_arch");
+  ShellCommandId read_arch_cmd_id = add_openfpga_read_arch_command(shell,
-  /* Add an option '--file' in short '-f'*/
+                                                                   openfpga_setup_cmd_class);
-  CommandOptionId read_arch_opt_file = shell_cmd_read_arch.add_option("file", true, "file path to the architecture XML");
-  shell_cmd_read_arch.set_option_short_name(read_arch_opt_file, "f");
-  shell_cmd_read_arch.set_option_require_value(read_arch_opt_file, openfpga::OPT_STRING);
-  
-  /* Add command 'read_openfpga_arch' to the Shell */
-  ShellCommandId shell_cmd_read_arch_id = shell.add_command(shell_cmd_read_arch, "read OpenFPGA architecture file");
-  shell.set_command_class(shell_cmd_read_arch_id, openfpga_setup_cmd_class);
-  shell.set_command_execute_function(shell_cmd_read_arch_id, read_arch);
 
   /******************************** 
    * Command 'write_openfpga_arch' 
    */
-  Command shell_cmd_write_arch("write_openfpga_arch");
+  std::vector<ShellCommandId> write_arch_dependent_cmds(1, read_arch_cmd_id);
-  /* Add an option '--file' in short '-f'*/
+  add_openfpga_write_arch_command(shell,
-  CommandOptionId write_arch_opt_file = shell_cmd_write_arch.add_option("file", true, "file path to the architecture XML");
+                                  openfpga_setup_cmd_class,
-  shell_cmd_write_arch.set_option_short_name(write_arch_opt_file, "f");
+                                  write_arch_dependent_cmds);
-  shell_cmd_write_arch.set_option_require_value(write_arch_opt_file, openfpga::OPT_STRING);
-
-  /* Add command 'write_openfpga_arch' to the Shell */
-  ShellCommandId shell_cmd_write_arch_id = shell.add_command(shell_cmd_write_arch, "write OpenFPGA architecture file");
-  shell.set_command_class(shell_cmd_write_arch_id, openfpga_setup_cmd_class);
-  shell.set_command_const_execute_function(shell_cmd_write_arch_id, write_arch);
-  /* The 'write_openfpga_arch' command should NOT be executed before 'read_openfpga_arch' */
-  shell.set_command_dependency(shell_cmd_write_arch_id, std::vector<ShellCommandId>(1, shell_cmd_read_arch_id));
 
   /******************************** 
    * Command 'link_openfpga_arch' 
    */
-  Command shell_cmd_link_openfpga_arch("link_openfpga_arch");
+  std::vector<ShellCommandId> link_arch_dependent_cmds;
-  /* Add an option '--verbose' */
+  link_arch_dependent_cmds.push_back(read_arch_cmd_id);
-  shell_cmd_link_openfpga_arch.add_option("verbose", false, "Show verbose outputs");
+  link_arch_dependent_cmds.push_back(vpr_cmd_id);
-  
+  ShellCommandId link_arch_cmd_id = add_openfpga_link_arch_command(shell,
-  /* Add command 'link_openfpga_arch' to the Shell */
+                                                                   openfpga_setup_cmd_class,
-  ShellCommandId shell_cmd_link_openfpga_arch_id = shell.add_command(shell_cmd_link_openfpga_arch, "Bind OpenFPGA architecture to VPR");
+                                                                   link_arch_dependent_cmds);
-  shell.set_command_class(shell_cmd_link_openfpga_arch_id, openfpga_setup_cmd_class);
-  shell.set_command_execute_function(shell_cmd_link_openfpga_arch_id, link_arch);
-  /* The 'link_openfpga_arch' command should NOT be executed before 'read_openfpga_arch' and 'vpr' */
-  std::vector<ShellCommandId> cmd_dependency_link_openfpga_arch;
-  cmd_dependency_link_openfpga_arch.push_back(shell_cmd_read_arch_id);
-  cmd_dependency_link_openfpga_arch.push_back(shell_cmd_vpr_id);
-  shell.set_command_dependency(shell_cmd_link_openfpga_arch_id, cmd_dependency_link_openfpga_arch);
-
   /******************************************* 
    * Command 'check_netlist_naming_conflict'
    */ 
-  Command shell_cmd_check_netlist_naming_conflict("check_netlist_naming_conflict");
+  std::vector<ShellCommandId> nlist_naming_dependent_cmds;
-  /* Add an option '--fix' */
+  nlist_naming_dependent_cmds.push_back(vpr_cmd_id);
-  shell_cmd_check_netlist_naming_conflict.add_option("fix", false, "Apply correction to any conflicts found");
+  add_openfpga_check_netlist_naming_conflict_command(shell,
-  /* Add an option '--report' */
+                                                     openfpga_setup_cmd_class,
-  CommandOptionId check_netlist_opt_rpt = shell_cmd_check_netlist_naming_conflict.add_option("report", false, "Output a report file about what any correction applied");
+                                                     nlist_naming_dependent_cmds);
-  shell_cmd_check_netlist_naming_conflict.set_option_require_value(check_netlist_opt_rpt, openfpga::OPT_STRING);
-
-  /* Add command 'check_netlist_naming_conflict' to the Shell */
-  ShellCommandId shell_cmd_check_netlist_naming_conflict_id = shell.add_command(shell_cmd_check_netlist_naming_conflict, "Check any block/net naming in users' BLIF netlist violates the syntax of fabric generator");
-  shell.set_command_class(shell_cmd_check_netlist_naming_conflict_id, openfpga_setup_cmd_class);
-  shell.set_command_execute_function(shell_cmd_check_netlist_naming_conflict_id, check_netlist_naming_conflict);
-  /* The 'link_openfpga_arch' command should NOT be executed before 'vpr' */
-  std::vector<ShellCommandId> cmd_dependency_check_netlist_naming_conflict(1, shell_cmd_vpr_id);
-  shell.set_command_dependency(shell_cmd_link_openfpga_arch_id, cmd_dependency_check_netlist_naming_conflict);
 
   /******************************** 
    * Command 'pb_pin_fixup' 
    */
-  Command shell_cmd_pb_pin_fixup("pb_pin_fixup");
+  std::vector<ShellCommandId> pb_pin_fixup_dependent_cmds;
-  /* Add an option '--verbose' */
+  pb_pin_fixup_dependent_cmds.push_back(read_arch_cmd_id);
-  shell_cmd_pb_pin_fixup.add_option("verbose", false, "Show verbose outputs");
+  pb_pin_fixup_dependent_cmds.push_back(vpr_cmd_id);
-
+  add_openfpga_pb_pin_fixup_command(shell,
-  /* Add command 'pb_pin_fixup' to the Shell */
+                                    openfpga_setup_cmd_class,
-  ShellCommandId shell_cmd_pb_pin_fixup_id = shell.add_command(shell_cmd_pb_pin_fixup, "Fix up the packing results due to pin swapping during routing stage");
+                                    pb_pin_fixup_dependent_cmds);
-  shell.set_command_class(shell_cmd_pb_pin_fixup_id, openfpga_setup_cmd_class);
-  shell.set_command_execute_function(shell_cmd_pb_pin_fixup_id, pb_pin_fixup);
-  /* The 'pb_pin_fixup' command should NOT be executed before 'read_openfpga_arch' and 'vpr' */
-  std::vector<ShellCommandId> cmd_dependency_pb_pin_fixup;
-  cmd_dependency_pb_pin_fixup.push_back(shell_cmd_read_arch_id);
-  cmd_dependency_pb_pin_fixup.push_back(shell_cmd_vpr_id);
-  shell.set_command_dependency(shell_cmd_pb_pin_fixup_id, cmd_dependency_pb_pin_fixup);
 
   /******************************** 
    * Command 'lut_truth_table_fixup' 
    */
-  Command shell_cmd_lut_truth_table_fixup("lut_truth_table_fixup");
+  std::vector<ShellCommandId> lut_tt_fixup_dependent_cmds;
-  /* Add an option '--verbose' */
+  lut_tt_fixup_dependent_cmds.push_back(read_arch_cmd_id);
-  shell_cmd_lut_truth_table_fixup.add_option("verbose", false, "Show verbose outputs");
+  lut_tt_fixup_dependent_cmds.push_back(vpr_cmd_id);
-
+  add_openfpga_lut_truth_table_fixup_command(shell,
-  /* Add command 'lut_truth_table_fixup' to the Shell */
+                                             openfpga_setup_cmd_class,
-  ShellCommandId shell_cmd_lut_truth_table_fixup_id = shell.add_command(shell_cmd_lut_truth_table_fixup, "Fix up the truth table of Look-Up Tables due to pin swapping during packing stage");
+                                             lut_tt_fixup_dependent_cmds);
-  shell.set_command_class(shell_cmd_lut_truth_table_fixup_id, openfpga_setup_cmd_class);
-  shell.set_command_execute_function(shell_cmd_lut_truth_table_fixup_id, lut_truth_table_fixup);
-  /* The 'lut_truth_table_fixup' command should NOT be executed before 'read_openfpga_arch' and 'vpr' */
-  std::vector<ShellCommandId> cmd_dependency_lut_truth_table_fixup;
-  cmd_dependency_lut_truth_table_fixup.push_back(shell_cmd_read_arch_id);
-  cmd_dependency_lut_truth_table_fixup.push_back(shell_cmd_vpr_id);
-  shell.set_command_dependency(shell_cmd_lut_truth_table_fixup_id, cmd_dependency_lut_truth_table_fixup);
-
   /******************************** 
    * Command 'build_fabric' 
    */
-  Command shell_cmd_build_fabric("build_fabric");
+  std::vector<ShellCommandId> build_fabric_dependent_cmds;
-  /* Add an option '--verbose' */
+  build_fabric_dependent_cmds.push_back(link_arch_cmd_id);
-  shell_cmd_build_fabric.add_option("compress_routing", false, "Compress the number of unique routing modules by identifying the unique GSBs");
+  add_openfpga_build_fabric_command(shell,
-  shell_cmd_build_fabric.add_option("duplicate_grid_pin", false, "Duplicate the pins on the same side of a grid");
+                                    openfpga_setup_cmd_class,
-  shell_cmd_build_fabric.add_option("verbose", false, "Show verbose outputs");
+                                    build_fabric_dependent_cmds);
-
-  /* Add command 'compact_routing_hierarchy' to the Shell */
-  ShellCommandId shell_cmd_build_fabric_id = shell.add_command(shell_cmd_build_fabric, "Build the FPGA fabric in a graph of modules");
-  shell.set_command_class(shell_cmd_build_fabric_id, openfpga_setup_cmd_class);
-  shell.set_command_execute_function(shell_cmd_build_fabric_id, build_fabric);
-  /* The 'build_fabric' command should NOT be executed before 'link_openfpga_arch' */
-  std::vector<ShellCommandId> cmd_dependency_build_fabric;
-  cmd_dependency_build_fabric.push_back(shell_cmd_link_openfpga_arch_id);
-  shell.set_command_dependency(shell_cmd_build_fabric_id, cmd_dependency_build_fabric);
 } 
 
 } /* end namespace openfpga */

openfpga/src/base/openfpga_verilog.cpp

@@ -15,7 +15,7 @@
 namespace openfpga {
 
 /********************************************************************
- * A wrapper function to call the fabric_verilog function of FPGA-Verilog 
+ * A wrapper function to call the fabric Verilog generator of FPGA-Verilog 
  *******************************************************************/
 void write_fabric_verilog(OpenfpgaContext& openfpga_ctx,
                           const Command& cmd, const CommandContext& cmd_context) {
@@ -26,9 +26,6 @@ void write_fabric_verilog(OpenfpgaContext& openfpga_ctx,
   CommandOptionId opt_include_signal_init = cmd.option("include_signal_init");
   CommandOptionId opt_support_icarus_simulator = cmd.option("support_icarus_simulator");
   CommandOptionId opt_print_user_defined_template = cmd.option("print_user_defined_template");
-  CommandOptionId opt_print_top_testbench = cmd.option("print_top_testbench");
-  CommandOptionId opt_print_formal_verification_top_netlist = cmd.option("print_formal_verification_top_netlist");
-  CommandOptionId opt_print_autocheck_top_testbench = cmd.option("print_autocheck_top_testbench");
   CommandOptionId opt_verbose = cmd.option("verbose");
 
   /* This is an intermediate data structure which is designed to modularize the FPGA-Verilog
@@ -41,9 +38,6 @@ void write_fabric_verilog(OpenfpgaContext& openfpga_ctx,
   options.set_include_signal_init(cmd_context.option_enable(cmd, opt_include_signal_init));
   options.set_support_icarus_simulator(cmd_context.option_enable(cmd, opt_support_icarus_simulator));
   options.set_print_user_defined_template(cmd_context.option_enable(cmd, opt_print_user_defined_template));
-  options.set_print_top_testbench(cmd_context.option_enable(cmd, opt_print_top_testbench));
-  options.set_print_formal_verification_top_netlist(cmd_context.option_enable(cmd, opt_print_formal_verification_top_netlist));
-  options.set_print_autocheck_top_testbench(cmd_context.option_value(cmd, opt_print_autocheck_top_testbench));
   options.set_verbose_output(cmd_context.option_enable(cmd, opt_verbose));
   options.set_compress_routing(openfpga_ctx.flow_manager().compress_routing());
   
@@ -56,4 +50,43 @@ void write_fabric_verilog(OpenfpgaContext& openfpga_ctx,
                       options);
 } 
 
+/********************************************************************
+ * A wrapper function to call the Verilog testbench generator of FPGA-Verilog 
+ *******************************************************************/
+void write_verilog_testbench(OpenfpgaContext& openfpga_ctx,
+                             const Command& cmd, const CommandContext& cmd_context) {
+
+  CommandOptionId opt_output_dir = cmd.option("file");
+  CommandOptionId opt_reference_benchmark = cmd.option("reference_benchmark_file_path");
+  CommandOptionId opt_print_top_testbench = cmd.option("print_top_testbench");
+  CommandOptionId opt_print_formal_verification_top_netlist = cmd.option("print_formal_verification_top_netlist");
+  CommandOptionId opt_print_preconfig_top_testbench = cmd.option("print_preconfig_top_testbench");
+  CommandOptionId opt_print_simulation_ini = cmd.option("print_simulation_ini");
+  CommandOptionId opt_verbose = cmd.option("verbose");
+
+  /* This is an intermediate data structure which is designed to modularize the FPGA-Verilog
+   * Keep it independent from any other outside data structures
+   */
+  VerilogTestbenchOption options;
+  options.set_output_directory(cmd_context.option_value(cmd, opt_output_dir));
+  options.set_reference_benchmark_file_path(cmd_context.option_value(cmd, opt_reference_benchmark));
+  options.set_print_formal_verification_top_netlist(cmd_context.option_enable(cmd, opt_print_formal_verification_top_netlist));
+  options.set_print_preconfig_top_testbench(cmd_context.option_enable(cmd, opt_print_preconfig_top_testbench));
+  options.set_print_top_testbench(cmd_context.option_enable(cmd, opt_print_top_testbench));
+  options.set_print_simulation_ini(cmd_context.option_value(cmd, opt_print_simulation_ini));
+  options.set_verbose_output(cmd_context.option_enable(cmd, opt_verbose));
+  
+  fpga_verilog_testbench(openfpga_ctx.module_graph(),
+                         openfpga_ctx.bitstream_manager(),
+                         openfpga_ctx.fabric_bitstream(),
+                         g_vpr_ctx.atom(),
+                         g_vpr_ctx.placement(),
+                         openfpga_ctx.io_location_map(),
+                         openfpga_ctx.vpr_netlist_annotation(),
+                         openfpga_ctx.arch().circuit_lib,
+                         openfpga_ctx.arch().sim_setting,
+                         openfpga_ctx.arch().config_protocol.type(),
+                         options);
+} 
+
 } /* end namespace openfpga */

openfpga/src/base/openfpga_verilog.h

@@ -18,6 +18,9 @@ namespace openfpga {
 void write_fabric_verilog(OpenfpgaContext& openfpga_ctx,
                           const Command& cmd, const CommandContext& cmd_context); 
 
+void write_verilog_testbench(OpenfpgaContext& openfpga_ctx,
+                             const Command& cmd, const CommandContext& cmd_context); 
+
 } /* end namespace openfpga */
 
 #endif

openfpga/src/base/openfpga_verilog_command.cpp

@@ -11,6 +11,98 @@
 /* begin namespace openfpga */
 namespace openfpga {
 
+/********************************************************************
+ * - Add a command to Shell environment: generate fabric Verilog 
+ * - Add associated options 
+ * - Add command dependency
+ *******************************************************************/
+static 
+void add_openfpga_write_fabric_verilog_command(openfpga::Shell<OpenfpgaContext>& shell,
+                                               const ShellCommandClassId& cmd_class_id,
+                                               const ShellCommandId& shell_cmd_build_fabric_id) {
+  Command shell_cmd("write_fabric_verilog");
+
+  /* Add an option '--file' in short '-f'*/
+  CommandOptionId output_opt = shell_cmd.add_option("file", true, "Specify the output directory for Verilog netlists");
+  shell_cmd.set_option_short_name(output_opt, "f");
+  shell_cmd.set_option_require_value(output_opt, openfpga::OPT_STRING);
+
+  /* Add an option '--explicit_port_mapping' */
+  shell_cmd.add_option("explicit_port_mapping", false, "Use explicit port mapping in Verilog netlists");
+
+  /* Add an option '--include_timing' */
+  shell_cmd.add_option("include_timing", false, "Enable timing annotation in Verilog netlists");
+
+  /* Add an option '--include_signal_init' */
+  shell_cmd.add_option("include_signal_init", false, "Initialize all the signals in Verilog netlists");
+
+  /* Add an option '--support_icarus_simulator' */
+  shell_cmd.add_option("support_icarus_simulator", false, "Fine-tune Verilog netlists to support icarus simulator");
+
+  /* Add an option '--print_user_defined_template' */
+  shell_cmd.add_option("print_user_defined_template", false, "Generate a template Verilog files for user-defined circuit models");
+
+  /* Add an option '--verbose' */
+  shell_cmd.add_option("verbose", false, "Enable verbose output");
+  
+  /* Add command 'write_fabric_verilog' to the Shell */
+  ShellCommandId shell_cmd_id = shell.add_command(shell_cmd, "generate Verilog netlists modeling full FPGA fabric");
+  shell.set_command_class(shell_cmd_id, cmd_class_id);
+  shell.set_command_execute_function(shell_cmd_id, write_fabric_verilog);
+
+  /* The 'build_fabric' command should NOT be executed before 'link_openfpga_arch' */
+  std::vector<ShellCommandId> cmd_dependency;
+  cmd_dependency.push_back(shell_cmd_build_fabric_id);
+  shell.set_command_dependency(shell_cmd_id, cmd_dependency);
+}
+
+/********************************************************************
+ * - Add a command to Shell environment: write Verilog testbench
+ * - Add associated options 
+ * - Add command dependency
+ *******************************************************************/
+static 
+void add_openfpga_write_verilog_testbench_command(openfpga::Shell<OpenfpgaContext>& shell,
+                                                  const ShellCommandClassId& cmd_class_id,
+                                                  const ShellCommandId& shell_cmd_build_fabric_id) {
+  Command shell_cmd("write_verilog_testbench");
+
+  /* Add an option '--file' in short '-f'*/
+  CommandOptionId output_opt = shell_cmd.add_option("file", true, "Specify the output directory for Verilog netlists");
+  shell_cmd.set_option_short_name(output_opt, "f");
+  shell_cmd.set_option_require_value(output_opt, openfpga::OPT_STRING);
+
+  /* Add an option '--reference_benchmark_file_path'*/
+  CommandOptionId ref_bm_opt = shell_cmd.add_option("reference_benchmark_file_path", true, "Specify the file path to the reference Verilog netlist");
+  shell_cmd.set_option_require_value(ref_bm_opt, openfpga::OPT_STRING);
+
+  /* Add an option '--print_top_testbench' */
+  shell_cmd.add_option("print_top_testbench", false, "Generate a full testbench for top-level fabric module with autocheck capability");
+
+  /* Add an option '--print_formal_verification_top_netlist' */
+  shell_cmd.add_option("print_formal_verification_top_netlist", false, "Generate a top-level module which can be used in formal verification");
+
+  /* Add an option '--print_preconfig_top_testbench' */
+  shell_cmd.add_option("print_preconfig_top_testbench", false, "Generate a pre-configured testbench for top-level fabric module with autocheck capability");
+
+  /* Add an option '--print_simulation_ini' */
+  CommandOptionId sim_ini_opt = shell_cmd.add_option("print_simulation_ini", false, "Generate a .ini file as an exchangeable file to enable HDL simulations");
+  shell_cmd.set_option_require_value(sim_ini_opt, openfpga::OPT_STRING);
+
+  /* Add an option '--verbose' */
+  shell_cmd.add_option("verbose", false, "Enable verbose output");
+  
+  /* Add command to the Shell */
+  ShellCommandId shell_cmd_id = shell.add_command(shell_cmd, "generate Verilog testbenches for full FPGA fabric");
+  shell.set_command_class(shell_cmd_id, cmd_class_id);
+  shell.set_command_execute_function(shell_cmd_id, write_verilog_testbench);
+
+  /* The command should NOT be executed before 'build_fabric' */
+  std::vector<ShellCommandId> cmd_dependency;
+  cmd_dependency.push_back(shell_cmd_build_fabric_id);
+  shell.set_command_dependency(shell_cmd_id, cmd_dependency);
+}
+
 void add_openfpga_verilog_commands(openfpga::Shell<OpenfpgaContext>& shell) {
   /* Get the unique id of 'build_fabric' command which is to be used in creating the dependency graph */
   const ShellCommandId& shell_cmd_build_fabric_id = shell.command(std::string("build_fabric"));
@@ -19,42 +111,18 @@ void add_openfpga_verilog_commands(openfpga::Shell<OpenfpgaContext>& shell) {
   ShellCommandClassId openfpga_verilog_cmd_class = shell.add_command_class("FPGA-Verilog");
 
   /******************************** 
-   * Command 'wirte_fabric_verilog' 
+   * Command 'write_fabric_verilog' 
    */
-  Command shell_cmd_write_fabric_verilog("write_fabric_verilog");
+  add_openfpga_write_fabric_verilog_command(shell,
-  /* Add an option '--file' in short '-f'*/
+                                            openfpga_verilog_cmd_class,
-  CommandOptionId fabric_verilog_output_opt = shell_cmd_write_fabric_verilog.add_option("file", true, "Specify the output directory for Verilog netlists");
+                                            shell_cmd_build_fabric_id);
-  shell_cmd_write_fabric_verilog.set_option_short_name(fabric_verilog_output_opt, "f");
-  shell_cmd_write_fabric_verilog.set_option_require_value(fabric_verilog_output_opt, openfpga::OPT_STRING);
-  /* Add an option '--explicit_port_mapping' */
-  shell_cmd_write_fabric_verilog.add_option("explicit_port_mapping", false, "Use explicit port mapping in Verilog netlists");
-  /* Add an option '--include_timing' */
-  shell_cmd_write_fabric_verilog.add_option("include_timing", false, "Enable timing annotation in Verilog netlists");
-  /* Add an option '--include_signal_init' */
-  shell_cmd_write_fabric_verilog.add_option("include_signal_init", false, "Initialize all the signals in Verilog netlists");
-  /* Add an option '--support_icarus_simulator' */
-  shell_cmd_write_fabric_verilog.add_option("support_icarus_simulator", false, "Fine-tune Verilog netlists to support icarus simulator");
-  /* Add an option '--print_user_defined_template' */
-  shell_cmd_write_fabric_verilog.add_option("print_user_defined_template", false, "Generate a template Verilog files for user-defined circuit models");
-  /* Add an option '--print_top_testbench' */
-  shell_cmd_write_fabric_verilog.add_option("print_top_testbench", false, "Generate a testbench for top-level fabric module");
-  /* Add an option '--print_formal_verification_top_netlist' */
-  shell_cmd_write_fabric_verilog.add_option("print_formal_verification_top_netlist", false, "Generate a top-level module which can be used in formal verification");
-  /* Add an option '--print_autocheck_top_testbench' */
-  CommandOptionId fabric_verilog_autocheck_tb_opt = shell_cmd_write_fabric_verilog.add_option("print_autocheck_top_testbench", false, "Generate a testbench for top-level fabric module with autocheck capability");
-  shell_cmd_write_fabric_verilog.set_option_require_value(fabric_verilog_autocheck_tb_opt, openfpga::OPT_STRING);
-  /* Add an option '--verbose' */
-  shell_cmd_write_fabric_verilog.add_option("verbose", false, "Enable verbose output");
 
-  /* Add command 'write_fabric_verilog' to the Shell */
+  /******************************** 
-  ShellCommandId shell_cmd_write_fabric_verilog_id = shell.add_command(shell_cmd_write_fabric_verilog, "generate Verilog netlists modeling full FPGA fabric");
+   * Command 'write_verilog_testbench' 
-  shell.set_command_class(shell_cmd_write_fabric_verilog_id, openfpga_verilog_cmd_class);
+   */
-  shell.set_command_execute_function(shell_cmd_write_fabric_verilog_id, write_fabric_verilog);
+  add_openfpga_write_verilog_testbench_command(shell,
-
+                                               openfpga_verilog_cmd_class,
-  /* The 'build_fabric' command should NOT be executed before 'link_openfpga_arch' */
+                                               shell_cmd_build_fabric_id);
-  std::vector<ShellCommandId> cmd_dependency_write_fabric_verilog;
-  cmd_dependency_write_fabric_verilog.push_back(shell_cmd_build_fabric_id);
-  shell.set_command_dependency(shell_cmd_write_fabric_verilog_id, cmd_dependency_write_fabric_verilog);
 } 
 
 } /* end namespace openfpga */

openfpga/src/fabric/build_device_module.cpp

@@ -26,7 +26,8 @@ namespace openfpga {
  * The main function to be called for building module graphs 
  * for a FPGA fabric
  *******************************************************************/
-ModuleManager build_device_module_graph(const DeviceContext& vpr_device_ctx,
+ModuleManager build_device_module_graph(IoLocationMap& io_location_map,
+                                        const DeviceContext& vpr_device_ctx,
                                         const OpenfpgaContext& openfpga_ctx,
                                         const bool& compress_routing,
                                         const bool& duplicate_grid_pin,
@@ -97,7 +98,8 @@ ModuleManager build_device_module_graph(const DeviceContext& vpr_device_ctx,
   }
 
   /* Build FPGA fabric top-level module */
-  build_top_module(module_manager, openfpga_ctx.arch().circuit_lib, 
+  build_top_module(module_manager, io_location_map,
+                   openfpga_ctx.arch().circuit_lib, 
                    vpr_device_ctx.grid,
                    vpr_device_ctx.rr_graph,
                    openfpga_ctx.device_rr_gsb(), 

openfpga/src/fabric/build_device_module.h

@@ -14,7 +14,8 @@
 /* begin namespace openfpga */
 namespace openfpga {
 
-ModuleManager build_device_module_graph(const DeviceContext& vpr_device_ctx,
+ModuleManager build_device_module_graph(IoLocationMap& io_location_map,
+                                        const DeviceContext& vpr_device_ctx,
                                         const OpenfpgaContext& openfpga_ctx,
                                         const bool& compress_routing,
                                         const bool& duplicate_grid_pin,

openfpga/src/fabric/build_grid_modules.cpp

@@ -1103,7 +1103,6 @@ void build_grid_modules(ModuleManager& module_manager,
                                    duplicate_grid_pin,
                                    verbose);
       } 
-      continue;
     } else {
       /* For CLB and heterogenenous blocks */
       build_physical_tile_module(module_manager, circuit_lib,

openfpga/src/fabric/build_top_module.cpp

@@ -89,6 +89,7 @@ size_t add_top_module_grid_instance(ModuleManager& module_manager,
 static 
 vtr::Matrix<size_t> add_top_module_grid_instances(ModuleManager& module_manager,
                                                   const ModuleId& top_module,
+                                                  IoLocationMap& io_location_map,
                                                   const DeviceGrid& grids) {
   /* Reserve an array for the instance ids */
   vtr::Matrix<size_t> grid_instance_ids({grids.width(), grids.height()}); 
@@ -142,6 +143,7 @@ vtr::Matrix<size_t> add_top_module_grid_instances(ModuleManager& module_manager,
   }
 
   /* Add instances of I/O grids to top_module */
+  size_t io_counter = 0;
   for (const e_side& io_side : io_sides) {
     for (const vtr::Point<size_t>& io_coordinate : io_coordinates[io_side]) {
       /* Bypass EMPTY grid */
@@ -157,6 +159,21 @@ vtr::Matrix<size_t> add_top_module_grid_instances(ModuleManager& module_manager,
       VTR_ASSERT(true == is_io_type(grids[io_coordinate.x()][io_coordinate.y()].type));
       /* Add a grid module to top_module*/
       grid_instance_ids[io_coordinate.x()][io_coordinate.y()] = add_top_module_grid_instance(module_manager, top_module, grids[io_coordinate.x()][io_coordinate.y()].type, io_side, io_coordinate);
+
+      /* MUST DO: register in io location mapping!
+       * I/O location mapping is a critical look-up for testbench generators
+       * As we add the I/O grid instances to top module by following order:
+       * TOP -> RIGHT -> BOTTOM -> LEFT
+       * The I/O index will increase in this way as well.
+       * This organization I/O indices is also consistent to the way 
+       * that GPIOs are wired in function connect_gpio_module()
+       *
+       * Note: if you change the GPIO function, you should update here as well!
+       */
+      for (int z = 0; z < grids[io_coordinate.x()][io_coordinate.y()].type->capacity; ++z) {
+        io_location_map.set_io_index(io_coordinate.x(), io_coordinate.y(), z, io_counter);
+        io_counter++;
+      }
     }
   }
 
@@ -280,6 +297,7 @@ vtr::Matrix<size_t> add_top_module_connection_block_instances(ModuleManager& mod
  * 5. Add module nets/submodules to connect configuration ports
  *******************************************************************/
 void build_top_module(ModuleManager& module_manager,
+                      IoLocationMap& io_location_map,
                       const CircuitLibrary& circuit_lib,
                       const DeviceGrid& grids,
                       const RRGraph& rr_graph,
@@ -301,7 +319,7 @@ void build_top_module(ModuleManager& module_manager,
  
   /* Add sub modules, which are grid, SB and CBX/CBY modules as instances */
   /* Add all the grids across the fabric */
-  vtr::Matrix<size_t> grid_instance_ids = add_top_module_grid_instances(module_manager, top_module, grids);
+  vtr::Matrix<size_t> grid_instance_ids = add_top_module_grid_instances(module_manager, top_module, io_location_map, grids);
   /* Add all the SBs across the fabric */
   vtr::Matrix<size_t> sb_instance_ids = add_top_module_switch_block_instances(module_manager, top_module, device_rr_gsb, compact_routing_hierarchy);
   /* Add all the CBX and CBYs across the fabric */

openfpga/src/fabric/build_top_module.h

@@ -14,6 +14,7 @@
 #include "tile_direct.h"
 #include "arch_direct.h"
 #include "module_manager.h"
+#include "io_location_map.h"
 
 /********************************************************************
  * Function declaration
@@ -23,6 +24,7 @@
 namespace openfpga {
 
 void build_top_module(ModuleManager& module_manager,
+                      IoLocationMap& io_location_map,
                       const CircuitLibrary& circuit_lib,
                       const DeviceGrid& grids,
                       const RRGraph& rr_graph,

openfpga/src/fpga_bitstream/build_grid_bitstream.cpp

@@ -168,7 +168,8 @@ void build_physical_block_pin_interc_bitstream(BitstreamManager& bitstream_manag
       for (t_pb_graph_pin* src_pb_graph_pin : pb_graph_pin_inputs(des_pb_graph_pin, cur_interc)) {
         const PhysicalPbId& src_pb_id = physical_pb.find_pb(src_pb_graph_pin->parent_node);
         /* If the src pb id is not valid, we bypass it */
-        if ( (true != physical_pb.valid_pb_id(src_pb_id))
+        if ( (true == physical_pb.valid_pb_id(src_pb_id))
+          && (AtomNetId::INVALID() != physical_pb.pb_graph_pin_atom_net(des_pb_id, des_pb_graph_pin))
           && (physical_pb.pb_graph_pin_atom_net(src_pb_id, src_pb_graph_pin) == physical_pb.pb_graph_pin_atom_net(des_pb_id, des_pb_graph_pin))) {
           break;
         }

openfpga/src/fpga_sdc/analysis_sdc_option.cpp

@@ -0,0 +1,39 @@
+/********************************************************************
+ * Member functions for a data structure which includes all the options for the SDC generator
+ ********************************************************************/
+#include "analysis_sdc_option.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * Public Constructors
+ ********************************************************************/
+AnalysisSdcOption::AnalysisSdcOption(const std::string& sdc_dir) {
+  sdc_dir_ = sdc_dir;
+  generate_sdc_analysis_ = false;
+}
+
+/********************************************************************
+ * Public accessors
+ ********************************************************************/
+std::string AnalysisSdcOption::sdc_dir() const {
+  return sdc_dir_;
+}
+
+bool AnalysisSdcOption::generate_sdc_analysis() const {
+  return generate_sdc_analysis_;
+}
+
+/********************************************************************
+ * Public mutators
+ ********************************************************************/
+void AnalysisSdcOption::set_sdc_dir(const std::string& sdc_dir) {
+  sdc_dir_ = sdc_dir;
+}
+
+void AnalysisSdcOption::set_generate_sdc_analysis(const bool& generate_sdc_analysis) {
+  generate_sdc_analysis_ = generate_sdc_analysis;
+}
+
+} /* end namespace openfpga */

openfpga/src/fpga_sdc/analysis_sdc_option.h

@@ -0,0 +1,30 @@
+#ifndef ANALYSIS_SDC_OPTION_H
+#define ANALYSIS_SDC_OPTION_H
+
+/********************************************************************
+ * A data structure to include all the options for the SDC generator
+ * in purpose of analyzing users' implementations
+ ********************************************************************/
+
+#include <string>
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+class AnalysisSdcOption {
+  public: /* Public Constructors */
+    AnalysisSdcOption(const std::string& sdc_dir);
+  public: /* Public accessors */
+    std::string sdc_dir() const;
+    bool generate_sdc_analysis() const;
+  public: /* Public mutators */
+    void set_sdc_dir(const std::string& sdc_dir);
+    void set_generate_sdc_analysis(const bool& generate_sdc_analysis);
+  private: /* Internal data */
+    std::string sdc_dir_;
+    bool generate_sdc_analysis_; 
+};
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/fpga_sdc/pnr_sdc_grid_writer.cpp

@@ -0,0 +1,346 @@
+/********************************************************************
+ * This file includes functions that print SDC (Synopsys Design Constraint) 
+ * files in physical design tools, i.e., Place & Route (PnR) tools
+ * The SDC files are used to constrain the physical design for each grid
+ * (CLBs, heterogeneous blocks etc.)
+ *
+ * Note that this is different from the SDC to constrain VPR Place&Route
+ * engine! These SDCs are designed for PnR to generate FPGA layouts!!!
+ *******************************************************************/
+#include <ctime>
+#include <fstream>
+
+/* Headers from vtrutil library */
+#include "vtr_log.h"
+#include "vtr_assert.h"
+#include "vtr_time.h"
+
+/* Headers from openfpgautil library */
+#include "openfpga_port.h"
+#include "openfpga_digest.h"
+#include "openfpga_side_manager.h"
+
+
+#include "openfpga_interconnect_types.h"
+#include "vpr_utils.h"
+#include "mux_utils.h"
+
+#include "openfpga_reserved_words.h"
+#include "openfpga_naming.h"
+#include "pb_type_utils.h"
+#include "pb_graph_utils.h"
+
+#include "sdc_writer_naming.h"
+#include "sdc_writer_utils.h"
+#include "pnr_sdc_grid_writer.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * Print pin-to-pin timing constraints for a given interconnection
+ * at an output port of a pb_graph node
+ *******************************************************************/
+static 
+void print_pnr_sdc_constrain_pb_pin_interc_timing(std::fstream& fp,
+                                                  const ModuleManager& module_manager,
+                                                  const ModuleId& parent_module,
+                                                  t_pb_graph_pin* des_pb_graph_pin,
+                                                  t_mode* physical_mode) {
+
+  /* Validate file stream */ 
+  valid_file_stream(fp);
+
+  /* 1. identify pin interconnection type, 
+   * 2. Identify the number of fan-in (Consider interconnection edges of only selected mode)
+   * 3. Print SDC timing constraints
+   */
+  t_interconnect* cur_interc = pb_graph_pin_interc(des_pb_graph_pin, physical_mode);
+  size_t fan_in = pb_graph_pin_inputs(des_pb_graph_pin, cur_interc).size();
+  if ((nullptr == cur_interc) || (0 == fan_in)) { 
+    /* No interconnection matched */
+    return;
+  }
+
+  /* Print pin-to-pin SDC contraint here */ 
+  /* For more than one mode defined, the direct interc has more than one input_edge ,
+   * We need to find which edge is connected the pin we want
+   */
+  for (int iedge = 0; iedge < des_pb_graph_pin->num_input_edges; iedge++) {
+    if (cur_interc != des_pb_graph_pin->input_edges[iedge]->interconnect) {
+      continue;
+    }
+
+    /* Source pin, node, pb_type*/
+    t_pb_graph_pin* src_pb_graph_pin = des_pb_graph_pin->input_edges[iedge]->input_pins[0];
+    t_pb_graph_node* src_pb_graph_node = src_pb_graph_pin->parent_node;
+    /* Des pin, node, pb_type */
+    t_pb_graph_node* des_pb_graph_node  = des_pb_graph_pin->parent_node;
+
+    /* Find the src module in module manager */
+    std::string src_module_name = generate_physical_block_module_name(src_pb_graph_pin->parent_node->pb_type);
+    ModuleId src_module = module_manager.find_module(src_module_name);
+    VTR_ASSERT(true == module_manager.valid_module_id(src_module));
+
+    ModulePortId src_module_port_id = module_manager.find_module_port(src_module, generate_pb_type_port_name(src_pb_graph_pin->port));
+    VTR_ASSERT(true == module_manager.valid_module_port_id(src_module, src_module_port_id));
+
+    /* Generate the name of the des instance name 
+     * If des module is not the parent module, it is a child module.
+     * We should find the instance id 
+     */
+    std::string src_instance_name = src_module_name; 
+    if (parent_module != src_module) {
+      src_instance_name = module_manager.module_name(parent_module) + std::string("/"); 
+      /* Instance id is actually the placement index */
+      size_t instance_id = src_pb_graph_node->placement_index;
+      if (true == module_manager.instance_name(parent_module, src_module, instance_id).empty()) {  
+        src_instance_name += src_module_name; 
+        src_instance_name += "_";
+        src_instance_name += std::to_string(instance_id);
+        src_instance_name += "_";
+      } else {
+        src_instance_name += module_manager.instance_name(parent_module, src_module, instance_id);  
+      }
+    }
+
+    /* Generate src port information */
+    BasicPort src_port = module_manager.module_port(src_module, src_module_port_id);
+    src_port.set_width(src_pb_graph_pin->pin_number, src_pb_graph_pin->pin_number);
+
+    /* Find the des module in module manager */
+    std::string des_module_name = generate_physical_block_module_name(des_pb_graph_pin->parent_node->pb_type);
+    ModuleId des_module = module_manager.find_module(des_module_name);
+    VTR_ASSERT(true == module_manager.valid_module_id(des_module));
+    ModulePortId des_module_port_id = module_manager.find_module_port(des_module, generate_pb_type_port_name(des_pb_graph_pin->port));
+    VTR_ASSERT(true == module_manager.valid_module_port_id(des_module, des_module_port_id));
+
+    /* Generate the name of the des instance name 
+     * If des module is not the parent module, it is a child module.
+     * We should find the instance id 
+     */
+    std::string des_instance_name = des_module_name; 
+    if (parent_module != des_module) {
+      des_instance_name = module_manager.module_name(parent_module) + std::string("/"); 
+      /* Instance id is actually the placement index */
+      size_t instance_id = des_pb_graph_node->placement_index;
+      if (true == module_manager.instance_name(parent_module, des_module, instance_id).empty()) {  
+        des_instance_name += des_module_name; 
+        des_instance_name += "_";
+        des_instance_name += std::to_string(instance_id);
+        des_instance_name += "_";
+      } else {
+        des_instance_name += module_manager.instance_name(parent_module, des_module, instance_id);  
+      }
+    }
+
+    /* Generate des port information */
+    BasicPort des_port = module_manager.module_port(des_module, des_module_port_id);
+    des_port.set_width(des_pb_graph_pin->pin_number, des_pb_graph_pin->pin_number);
+
+    /* Print a SDC timing constraint */
+    print_pnr_sdc_constrain_max_delay(fp, 
+                                      src_instance_name, 
+                                      generate_sdc_port(src_port),
+                                      des_instance_name, 
+                                      generate_sdc_port(des_port),
+                                      des_pb_graph_pin->input_edges[iedge]->delay_max);
+  }
+}
+
+/********************************************************************
+ * Print port-to-port timing constraints which source from
+ * an output port of a pb_graph node
+ *******************************************************************/
+static 
+void print_pnr_sdc_constrain_pb_interc_timing(std::fstream& fp,
+                                              const ModuleManager& module_manager,
+                                              const ModuleId& parent_module,
+                                              t_pb_graph_node* des_pb_graph_node,
+                                              const e_circuit_pb_port_type& pb_port_type,
+                                              t_mode* physical_mode) {
+  /* Validate file stream */ 
+  valid_file_stream(fp);
+
+  switch (pb_port_type) {
+  case CIRCUIT_PB_PORT_INPUT: {
+    for (int iport = 0; iport < des_pb_graph_node->num_input_ports; ++iport) {
+      for (int ipin = 0; ipin < des_pb_graph_node->num_input_pins[iport]; ++ipin) {
+        /* If this is a idle block, we set 0 to the selected edge*/
+        /* Get the selected edge of current pin*/
+        print_pnr_sdc_constrain_pb_pin_interc_timing(fp, 
+                                                     module_manager, parent_module, 
+                                                     &(des_pb_graph_node->input_pins[iport][ipin]),
+                                                     physical_mode);
+      }
+    }
+    break;
+  }
+  case CIRCUIT_PB_PORT_OUTPUT: {
+    for (int iport = 0; iport < des_pb_graph_node->num_output_ports; ++iport) {
+      for (int ipin = 0; ipin < des_pb_graph_node->num_output_pins[iport]; ++ipin) {
+        print_pnr_sdc_constrain_pb_pin_interc_timing(fp,
+                                                     module_manager, parent_module, 
+                                                     &(des_pb_graph_node->output_pins[iport][ipin]),
+                                                     physical_mode);
+      }
+    }
+    break;
+  }
+  case CIRCUIT_PB_PORT_CLOCK: {
+    /* Do NOT constrain clock here, it should be handled by Clock Tree Synthesis */
+    break;
+  }
+  default:
+   VTR_LOGF_ERROR(__FILE__, __LINE__,
+                  "Invalid pb port type!\n");
+    exit(1);
+  }
+}
+
+/********************************************************************
+ * This function will generate a SDC file for each pb_type,
+ * constraining the pin-to-pin timing between 
+ * 1. input port of parent_pb_graph_node and input port of child_pb_graph_nodes
+ * 2. output port of parent_pb_graph_node and output port of child_pb_graph_nodes
+ * 3. output port of child_pb_graph_node and input port of child_pb_graph_nodes
+ *******************************************************************/
+static 
+void print_pnr_sdc_constrain_pb_graph_node_timing(const std::string& sdc_dir,
+                                                  const ModuleManager& module_manager,
+                                                  t_pb_graph_node* parent_pb_graph_node,
+                                                  t_mode* physical_mode) {
+
+  /* Get the pb_type definition related to the node */
+  t_pb_type* physical_pb_type = parent_pb_graph_node->pb_type; 
+  std::string pb_module_name = generate_physical_block_module_name(physical_pb_type);
+
+  /* Find the pb module in module manager */
+  ModuleId pb_module = module_manager.find_module(pb_module_name);
+  VTR_ASSERT(true == module_manager.valid_module_id(pb_module));
+
+  /* Create the file name for SDC */
+  std::string sdc_fname(sdc_dir + pb_module_name + std::string(SDC_FILE_NAME_POSTFIX));
+
+  /* Create the file stream */
+  std::fstream fp;
+  fp.open(sdc_fname, std::fstream::out | std::fstream::trunc);
+
+  check_file_stream(sdc_fname.c_str(), fp);
+
+  /* Generate the descriptions*/
+  print_sdc_file_header(fp, std::string("Timing constraints for Grid " + pb_module_name + " in PnR"));
+
+  /* We check output_pins of cur_pb_graph_node and its the input_edges
+   * Built the interconnections between outputs of cur_pb_graph_node and outputs of child_pb_graph_node
+   *   child_pb_graph_node.output_pins -----------------> cur_pb_graph_node.outpins
+   *                                        /|\
+   *                                         |
+   *                         input_pins,   edges,       output_pins
+   */ 
+  print_pnr_sdc_constrain_pb_interc_timing(fp,
+                                           module_manager, pb_module,
+                                           parent_pb_graph_node, 
+                                           CIRCUIT_PB_PORT_OUTPUT,
+                                           physical_mode);
+  
+  /* We check input_pins of child_pb_graph_node and its the input_edges
+   * Built the interconnections between inputs of cur_pb_graph_node and inputs of child_pb_graph_node
+   *   cur_pb_graph_node.input_pins -----------------> child_pb_graph_node.input_pins
+   *                                        /|\
+   *                                         |
+   *                         input_pins,   edges,       output_pins
+   */ 
+  for (int ipb = 0; ipb < physical_mode->num_pb_type_children; ++ipb) {
+    for (int jpb = 0; jpb < physical_mode->pb_type_children[ipb].num_pb; ++jpb) {
+      t_pb_graph_node* child_pb_graph_node = &(parent_pb_graph_node->child_pb_graph_nodes[physical_mode->index][ipb][jpb]);
+      /* For each child_pb_graph_node input pins*/
+      print_pnr_sdc_constrain_pb_interc_timing(fp,
+                                               module_manager, pb_module,
+                                               child_pb_graph_node, 
+                                               CIRCUIT_PB_PORT_INPUT,
+                                               physical_mode);
+      /* Do NOT constrain clock here, it should be handled by Clock Tree Synthesis */
+    }
+  }
+  
+  /* Close file handler */
+  fp.close();
+}
+
+/********************************************************************
+ * Recursively print SDC timing constraints for a pb_type
+ * This function will generate a SDC file for each pb_type,
+ * constraining the pin-to-pin timing
+ *******************************************************************/
+static 
+void rec_print_pnr_sdc_constrain_pb_graph_timing(const std::string& sdc_dir,
+                                                 const ModuleManager& module_manager,
+                                                 const VprDeviceAnnotation& device_annotation,
+                                                 t_pb_graph_node* parent_pb_graph_node) {
+  /* Validate pb_graph node */
+  if (nullptr == parent_pb_graph_node) {
+    VTR_LOGF_ERROR(__FILE__, __LINE__,
+                   "Invalid parent_pb_graph_node.\n");
+    exit(1);
+  }
+
+  /* Get the pb_type */
+  t_pb_type* parent_pb_type = parent_pb_graph_node->pb_type;
+
+  /* No need to constrain the primitive node */
+  if (true == is_primitive_pb_type(parent_pb_type)) {
+    return;
+  }
+
+  /* Note we only go through the graph through the physical modes. 
+   * which we build the modules 
+   */
+  t_mode* physical_mode = device_annotation.physical_mode(parent_pb_type);  
+
+  /* Write a SDC file for this pb_type */
+  print_pnr_sdc_constrain_pb_graph_node_timing(sdc_dir,
+                                               module_manager,
+                                               parent_pb_graph_node,
+                                               physical_mode);
+
+  /* Go recursively to the lower level in the pb_graph
+   * Note that we assume a full hierarchical P&R, we will only visit pb_graph_node of unique pb_type 
+   */
+  for (int ipb = 0; ipb < physical_mode->num_pb_type_children; ++ipb) {
+    rec_print_pnr_sdc_constrain_pb_graph_timing(sdc_dir, module_manager, 
+                                                device_annotation,
+                                                &(parent_pb_graph_node->child_pb_graph_nodes[physical_mode->index][ipb][0]));
+  }
+}
+
+/********************************************************************
+ * Top-level function to print timing constraints for pb_types
+ *******************************************************************/
+void print_pnr_sdc_constrain_grid_timing(const std::string& sdc_dir,
+                                         const DeviceContext& device_ctx,
+                                         const VprDeviceAnnotation& device_annotation,
+                                         const ModuleManager& module_manager) {
+
+  /* Start time count */
+  vtr::ScopedStartFinishTimer timer("Write SDC for constraining grid timing for P&R flow");
+
+  for (const t_physical_tile_type& physical_tile : device_ctx.physical_tile_types) {
+    /* Bypass empty type or nullptr */
+    if (true == is_empty_type(&physical_tile)) {
+      continue;
+    } else {
+      VTR_ASSERT(1 == physical_tile.equivalent_sites.size());
+      t_pb_graph_node* pb_graph_head = physical_tile.equivalent_sites[0]->pb_graph_head;
+      if (nullptr == pb_graph_head) {
+        continue;
+      }
+      /* Special for I/O block, generate one module for each border side */
+      rec_print_pnr_sdc_constrain_pb_graph_timing(sdc_dir, module_manager, 
+                                                  device_annotation,
+                                                  pb_graph_head);
+    }
+  }
+}
+
+} /* end namespace openfpga */

openfpga/src/fpga_sdc/pnr_sdc_grid_writer.h

@@ -0,0 +1,27 @@
+#ifndef PNR_SDC_GRID_WRITER_H
+#define PNR_SDC_GRID_WRITER_H
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+#include <string>
+#include <vector>
+#include "vpr_context.h"
+#include "vpr_device_annotation.h"
+#include "module_manager.h"
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+void print_pnr_sdc_constrain_grid_timing(const std::string& sdc_dir,
+                                         const DeviceContext& device_ctx,
+                                         const VprDeviceAnnotation& device_annotation,
+                                         const ModuleManager& module_manager);
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/fpga_sdc/pnr_sdc_option.cpp

@@ -0,0 +1,113 @@
+/********************************************************************
+ * Member functions for a data structure which includes all the options for the SDC generator
+ ********************************************************************/
+#include "pnr_sdc_option.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * Public Constructors
+ ********************************************************************/
+PnrSdcOption::PnrSdcOption(const std::string& sdc_dir) {
+  sdc_dir_ = sdc_dir;
+  constrain_global_port_ = false;
+  constrain_grid_ = false;
+  constrain_sb_ = false;
+  constrain_cb_ = false;
+  constrain_configurable_memory_outputs_ = false;
+  constrain_routing_multiplexer_outputs_ = false;
+  constrain_switch_block_outputs_ = false;
+}
+
+/********************************************************************
+ * Public accessors
+ ********************************************************************/
+std::string PnrSdcOption::sdc_dir() const {
+  return sdc_dir_;
+}
+
+bool PnrSdcOption::generate_sdc_pnr() const {
+  return constrain_global_port_ 
+      || constrain_grid_
+      || constrain_sb_
+      || constrain_cb_
+      || constrain_configurable_memory_outputs_
+      || constrain_routing_multiplexer_outputs_
+      || constrain_switch_block_outputs_;
+}
+
+bool PnrSdcOption::constrain_global_port() const {
+  return constrain_global_port_;
+}
+
+bool PnrSdcOption::constrain_grid() const {
+  return constrain_grid_;
+}
+
+bool PnrSdcOption::constrain_sb() const {
+  return constrain_sb_;
+}
+
+bool PnrSdcOption::constrain_cb() const {
+  return constrain_cb_;
+}
+
+bool PnrSdcOption::constrain_configurable_memory_outputs() const {
+  return constrain_configurable_memory_outputs_;
+}
+
+bool PnrSdcOption::constrain_routing_multiplexer_outputs() const {
+  return constrain_routing_multiplexer_outputs_;
+}
+
+bool PnrSdcOption::constrain_switch_block_outputs() const {
+  return constrain_switch_block_outputs_;
+}
+
+/********************************************************************
+ * Public mutators
+ ********************************************************************/
+void PnrSdcOption::set_sdc_dir(const std::string& sdc_dir) {
+  sdc_dir_ = sdc_dir;
+}
+
+void PnrSdcOption::set_generate_sdc_pnr(const bool& generate_sdc_pnr) {
+  constrain_global_port_ = generate_sdc_pnr;
+  constrain_grid_ = generate_sdc_pnr;
+  constrain_sb_ = generate_sdc_pnr;
+  constrain_cb_ = generate_sdc_pnr;
+  constrain_configurable_memory_outputs_ = generate_sdc_pnr;
+  constrain_routing_multiplexer_outputs_ = generate_sdc_pnr;
+  constrain_switch_block_outputs_ = generate_sdc_pnr;
+}
+
+void PnrSdcOption::set_constrain_global_port(const bool& constrain_global_port) {
+  constrain_global_port_ = constrain_global_port;
+}
+
+void PnrSdcOption::set_constrain_grid(const bool& constrain_grid) {
+  constrain_grid_ = constrain_grid;
+}
+
+void PnrSdcOption::set_constrain_sb(const bool& constrain_sb) {
+  constrain_sb_ = constrain_sb;
+}
+
+void PnrSdcOption::set_constrain_cb(const bool& constrain_cb) {
+  constrain_cb_ = constrain_cb;
+}
+
+void PnrSdcOption::set_constrain_configurable_memory_outputs(const bool& constrain_config_mem_outputs) {
+  constrain_configurable_memory_outputs_ = constrain_config_mem_outputs;
+}
+
+void PnrSdcOption::set_constrain_routing_multiplexer_outputs(const bool& constrain_routing_mux_outputs) {
+  constrain_routing_multiplexer_outputs_ = constrain_routing_mux_outputs;
+}
+
+void PnrSdcOption::set_constrain_switch_block_outputs(const bool& constrain_sb_outputs) {
+  constrain_switch_block_outputs_ = constrain_sb_outputs;
+}
+
+} /* end namespace openfpga */

openfpga/src/fpga_sdc/pnr_sdc_option.h

@@ -0,0 +1,50 @@
+#ifndef PNR_SDC_OPTION_H
+#define PNR_SDC_OPTION_H
+
+/********************************************************************
+ * A data structure to include all the options for the SDC generator
+ * in purpose of constraining physical design of FPGA fabric in back-end flow
+ ********************************************************************/
+
+#include <string>
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+class PnrSdcOption {
+  public: /* Public Constructors */
+    PnrSdcOption(const std::string& sdc_dir);
+  public: /* Public accessors */
+    std::string sdc_dir() const;
+    bool generate_sdc_pnr() const;
+    bool constrain_global_port() const;
+    bool constrain_grid() const;
+    bool constrain_sb() const;
+    bool constrain_cb() const;
+    bool constrain_configurable_memory_outputs() const;
+    bool constrain_routing_multiplexer_outputs() const;
+    bool constrain_switch_block_outputs() const;
+  public: /* Public mutators */
+    void set_sdc_dir(const std::string& sdc_dir);
+    void set_generate_sdc_pnr(const bool& generate_sdc_pnr);
+    void set_constrain_global_port(const bool& constrain_global_port);
+    void set_constrain_grid(const bool& constrain_grid);
+    void set_constrain_sb(const bool& constrain_sb);
+    void set_constrain_cb(const bool& constrain_cb);
+    void set_constrain_configurable_memory_outputs(const bool& constrain_config_mem_outputs);
+    void set_constrain_routing_multiplexer_outputs(const bool& constrain_routing_mux_outputs);
+    void set_constrain_switch_block_outputs(const bool& constrain_sb_outputs);
+  private: /* Internal data */
+    std::string sdc_dir_;
+    bool constrain_global_port_; 
+    bool constrain_grid_; 
+    bool constrain_sb_;
+    bool constrain_cb_;
+    bool constrain_configurable_memory_outputs_;
+    bool constrain_routing_multiplexer_outputs_;
+    bool constrain_switch_block_outputs_;
+};
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/fpga_sdc/pnr_sdc_routing_writer.cpp

@@ -0,0 +1,411 @@
+/********************************************************************
+ * This file includes functions that print SDC (Synopsys Design Constraint) 
+ * files in physical design tools, i.e., Place & Route (PnR) tools
+ * The SDC files are used to constrain the physical design for each routing modules
+ * in FPGA fabric, such as Switch Blocks (SBs) and Connection Blocks (CBs)
+ *
+ * Note that this is different from the SDC to constrain VPR Place&Route
+ * engine! These SDCs are designed for PnR to generate FPGA layouts!!!
+ *******************************************************************/
+#include <ctime>
+#include <fstream>
+
+/* Headers from vtrutil library */
+#include "vtr_log.h"
+#include "vtr_assert.h"
+#include "vtr_time.h"
+
+/* Headers from openfpgautil library */
+#include "openfpga_port.h"
+#include "openfpga_side_manager.h"
+#include "openfpga_digest.h"
+
+#include "mux_utils.h"
+
+#include "openfpga_naming.h"
+
+#include "openfpga_rr_graph_utils.h"
+#include "build_routing_module_utils.h"
+
+#include "sdc_writer_naming.h"
+#include "sdc_writer_utils.h"
+#include "pnr_sdc_routing_writer.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * Find the timing constraints between the inputs and outputs of a routing
+ * multiplexer in a Switch Block
+ *******************************************************************/
+static 
+float find_pnr_sdc_switch_tmax(const t_rr_switch_inf& switch_inf) {
+  return switch_inf.R * switch_inf.Cout + switch_inf.Tdel;
+}
+
+/********************************************************************
+ * Set timing constraints between the inputs and outputs of a routing
+ * multiplexer in a Switch Block
+ *******************************************************************/
+static 
+void print_pnr_sdc_constrain_sb_mux_timing(std::fstream& fp,
+                                           const ModuleManager& module_manager,
+                                           const ModuleId& sb_module, 
+                                           const RRGraph& rr_graph,
+                                           const RRGSB& rr_gsb,
+                                           const e_side& output_node_side,
+                                           const RRNodeId& output_rr_node) {
+  /* Validate file stream */
+  valid_file_stream(fp);
+
+  VTR_ASSERT(  ( CHANX == rr_graph.node_type(output_rr_node) )
+            || ( CHANY == rr_graph.node_type(output_rr_node) ));
+
+  /* Find the module port corresponding to the output rr_node */
+  ModulePortId module_output_port = find_switch_block_module_chan_port(module_manager, 
+                                                                       sb_module,
+                                                                       rr_graph, 
+                                                                       rr_gsb, 
+                                                                       output_node_side,
+                                                                       output_rr_node,
+                                                                       OUT_PORT);
+
+  /* Find the module port corresponding to the fan-in rr_nodes of the output rr_node */
+  std::vector<ModulePortId> module_input_ports = find_switch_block_module_input_ports(module_manager,
+                                                                                      sb_module, 
+                                                                                      rr_graph, 
+                                                                                      rr_gsb, 
+                                                                                      get_rr_graph_configurable_driver_nodes(rr_graph, output_rr_node));
+
+  /* Find timing constraints for each path (edge) */
+  std::map<ModulePortId, float> switch_delays;
+  size_t edge_counter = 0;
+  for (const RREdgeId& edge : rr_graph.node_configurable_in_edges(output_rr_node)) {
+    /* Get the switch delay */
+    const RRSwitchId& driver_switch = rr_graph.edge_switch(edge);
+    switch_delays[module_input_ports[edge_counter]] = find_pnr_sdc_switch_tmax(rr_graph.get_switch(driver_switch));
+    edge_counter++;
+  }
+
+  /* Find the starting points */
+  for (const ModulePortId& module_input_port : module_input_ports) {
+    /* Constrain a path */
+    print_pnr_sdc_constrain_port2port_timing(fp,
+                                             module_manager, 
+                                             sb_module, module_input_port, 
+                                             sb_module, module_output_port,
+                                             switch_delays[module_input_port]);
+  }
+}
+
+/********************************************************************
+ * Set timing constraints between the inputs and outputs of SBs, 
+ * which are connected by routing multiplexers with the given delays
+ * specified in architectural XML file
+ *
+ * To enable block by block timing constraining, we generate the SDC
+ * file for each unique SB module
+ *******************************************************************/
+static 
+void print_pnr_sdc_constrain_sb_timing(const std::string& sdc_dir,
+                                       const ModuleManager& module_manager,
+                                       const RRGraph& rr_graph,
+                                       const RRGSB& rr_gsb) {
+
+  /* Create the file name for Verilog netlist */
+  vtr::Point<size_t> gsb_coordinate(rr_gsb.get_sb_x(), rr_gsb.get_sb_y());
+  std::string sdc_fname(sdc_dir + generate_switch_block_module_name(gsb_coordinate) + std::string(SDC_FILE_NAME_POSTFIX));
+
+  /* Create the file stream */
+  std::fstream fp;
+  fp.open(sdc_fname, std::fstream::out | std::fstream::trunc);
+
+  /* Validate file stream */
+  check_file_stream(sdc_fname.c_str(), fp);
+
+  std::string sb_module_name = generate_switch_block_module_name(gsb_coordinate);
+  ModuleId sb_module = module_manager.find_module(sb_module_name);
+  VTR_ASSERT(true == module_manager.valid_module_id(sb_module));
+
+  /* Generate the descriptions*/
+  print_sdc_file_header(fp, std::string("Constrain timing of Switch Block " + sb_module_name + " for PnR"));
+
+  for (size_t side = 0; side < rr_gsb.get_num_sides(); ++side) {
+    SideManager side_manager(side);
+    for (size_t itrack = 0; itrack < rr_gsb.get_chan_width(side_manager.get_side()); ++itrack) {
+      const RRNodeId& chan_rr_node = rr_gsb.get_chan_node(side_manager.get_side(), itrack);
+      /* We only care the output port and it should indicate a SB mux */
+      if (OUT_PORT != rr_gsb.get_chan_node_direction(side_manager.get_side(), itrack)) { 
+        continue; 
+      }
+      /* Constrain thru wires */
+      if (false != rr_gsb.is_sb_node_passing_wire(rr_graph, side_manager.get_side(), itrack)) {
+        continue;
+      } 
+      /* This is a MUX, constrain all the paths from an input to an output */
+      print_pnr_sdc_constrain_sb_mux_timing(fp,
+                                            module_manager, sb_module, 
+                                            rr_graph,
+                                            rr_gsb,
+                                            side_manager.get_side(),
+                                            chan_rr_node);
+    }
+  }
+
+  /* Close file handler */
+  fp.close();
+}
+
+/********************************************************************
+ * Print SDC timing constraints for Switch blocks
+ * This function is designed for flatten routing hierarchy
+ *******************************************************************/
+void print_pnr_sdc_flatten_routing_constrain_sb_timing(const std::string& sdc_dir,
+                                                       const ModuleManager& module_manager,
+                                                       const RRGraph& rr_graph,
+                                                       const DeviceRRGSB& device_rr_gsb) {
+
+  /* Start time count */
+  vtr::ScopedStartFinishTimer timer("Write SDC for constrain Switch Block timing for P&R flow");
+
+  /* Get the range of SB array */
+  vtr::Point<size_t> sb_range = device_rr_gsb.get_gsb_range();
+  /* Go for each SB */
+  for (size_t ix = 0; ix < sb_range.x(); ++ix) {
+    for (size_t iy = 0; iy < sb_range.y(); ++iy) {
+      const RRGSB& rr_gsb = device_rr_gsb.get_gsb(ix, iy);
+      if (false == rr_gsb.is_sb_exist()) {
+        continue;
+      }
+      print_pnr_sdc_constrain_sb_timing(sdc_dir,
+                                        module_manager,
+                                        rr_graph,
+                                        rr_gsb);
+    }
+  }
+}
+
+/********************************************************************
+ * Print SDC timing constraints for Switch blocks
+ * This function is designed for compact routing hierarchy
+ *******************************************************************/
+void print_pnr_sdc_compact_routing_constrain_sb_timing(const std::string& sdc_dir,
+                                                       const ModuleManager& module_manager,
+                                                       const RRGraph& rr_graph,
+                                                       const DeviceRRGSB& device_rr_gsb) {
+
+  /* Start time count */
+  vtr::ScopedStartFinishTimer timer("Write SDC for constrain Switch Block timing for P&R flow");
+
+  for (size_t isb = 0; isb < device_rr_gsb.get_num_sb_unique_module(); ++isb) {
+    const RRGSB& rr_gsb = device_rr_gsb.get_sb_unique_module(isb);
+    if (false == rr_gsb.is_sb_exist()) {
+      continue;
+    }
+    print_pnr_sdc_constrain_sb_timing(sdc_dir,
+                                      module_manager,
+                                      rr_graph,
+                                      rr_gsb);
+  }
+}
+
+/********************************************************************
+ * Set timing constraints between the inputs and outputs of a routing
+ * multiplexer in a Connection Block
+ *******************************************************************/
+static 
+void print_pnr_sdc_constrain_cb_mux_timing(std::fstream& fp,
+                                           const ModuleManager& module_manager,
+                                           const ModuleId& cb_module, 
+                                           const RRGraph& rr_graph,
+                                           const RRGSB& rr_gsb,
+                                           const t_rr_type& cb_type,
+                                           const RRNodeId& output_rr_node) {
+  /* Validate file stream */
+  valid_file_stream(fp);
+
+  VTR_ASSERT(IPIN == rr_graph.node_type(output_rr_node));
+  
+  /* We have OPINs since we may have direct connections:
+   * These connections should be handled by other functions in the compact_netlist.c 
+   * So we just return here for OPINs 
+   */
+  if (0 == get_rr_graph_configurable_driver_nodes(rr_graph, output_rr_node).size()) {
+    return;
+  }
+
+  /* Find the module port corresponding to the output rr_node */
+  ModulePortId module_output_port = find_connection_block_module_ipin_port(module_manager, 
+                                                                           cb_module,
+                                                                           rr_graph, 
+                                                                           rr_gsb, 
+                                                                           output_rr_node);
+
+  /* Find the module port corresponding to the fan-in rr_nodes of the output rr_node */
+  std::vector<ModulePortId> module_input_ports = find_connection_block_module_input_ports(module_manager,
+                                                                                          cb_module, 
+                                                                                          rr_graph, 
+                                                                                          rr_gsb, 
+                                                                                          cb_type,
+                                                                                          get_rr_graph_configurable_driver_nodes(rr_graph, output_rr_node));
+
+  /* Find timing constraints for each path (edge) */
+  std::map<ModulePortId, float> switch_delays;
+  size_t edge_counter = 0;
+  for (const RREdgeId& edge : rr_graph.node_configurable_in_edges(output_rr_node)) {
+    /* Get the switch delay */
+    const RRSwitchId& driver_switch = rr_graph.edge_switch(edge);
+    switch_delays[module_input_ports[edge_counter]] = find_pnr_sdc_switch_tmax(rr_graph.get_switch(driver_switch));
+    edge_counter++;
+  }
+
+  /* Find the starting points */
+  for (const ModulePortId& module_input_port : module_input_ports) {
+    /* Constrain a path */
+    print_pnr_sdc_constrain_port2port_timing(fp,
+                                             module_manager, 
+                                             cb_module, module_input_port, 
+                                             cb_module, module_output_port,
+                                             switch_delays[module_input_port]);
+  }
+}
+
+
+/********************************************************************
+ * Print SDC timing constraints for a Connection block 
+ * This function is designed for compact routing hierarchy
+ *******************************************************************/
+static 
+void print_pnr_sdc_constrain_cb_timing(const std::string& sdc_dir,
+                                       const ModuleManager& module_manager,
+                                       const RRGraph& rr_graph,
+                                       const RRGSB& rr_gsb, 
+                                       const t_rr_type& cb_type) {
+  /* Create the netlist */
+  vtr::Point<size_t> gsb_coordinate(rr_gsb.get_cb_x(cb_type), rr_gsb.get_cb_y(cb_type));
+
+  /* Find the module name and create a SDC file for it */
+  std::string sdc_fname(sdc_dir + generate_connection_block_module_name(cb_type, gsb_coordinate) + std::string(SDC_FILE_NAME_POSTFIX));
+
+  /* Create the file stream */
+  std::fstream fp;
+  fp.open(sdc_fname, std::fstream::out | std::fstream::trunc);
+
+  /* Validate file stream */
+  check_file_stream(sdc_fname.c_str(), fp);
+
+  std::string cb_module_name = generate_connection_block_module_name(cb_type, gsb_coordinate); 
+  ModuleId cb_module = module_manager.find_module(cb_module_name);
+  VTR_ASSERT(true == module_manager.valid_module_id(cb_module));
+
+  /* Generate the descriptions*/
+  print_sdc_file_header(fp, std::string("Constrain timing of Connection Block " + cb_module_name + " for PnR"));
+
+  std::vector<enum e_side> cb_sides = rr_gsb.get_cb_ipin_sides(cb_type);
+
+  for (size_t side = 0; side < cb_sides.size(); ++side) {
+    enum e_side cb_ipin_side = cb_sides[side];
+    SideManager side_manager(cb_ipin_side);
+    for (size_t inode = 0; inode < rr_gsb.get_num_ipin_nodes(cb_ipin_side); ++inode) {
+      const RRNodeId& ipin_rr_node = rr_gsb.get_ipin_node(cb_ipin_side, inode);
+      print_pnr_sdc_constrain_cb_mux_timing(fp,
+                                            module_manager, cb_module, 
+                                            rr_graph, rr_gsb, cb_type,
+                                            ipin_rr_node);
+    }
+  }
+
+  /* Close file handler */
+  fp.close();
+}
+
+/********************************************************************
+ * Iterate over all the connection blocks in a device
+ * and print SDC file for each of them 
+ *******************************************************************/
+static 
+void print_pnr_sdc_flatten_routing_constrain_cb_timing(const std::string& sdc_dir,
+                                                       const ModuleManager& module_manager, 
+                                                       const RRGraph& rr_graph,
+                                                       const DeviceRRGSB& device_rr_gsb,
+                                                       const t_rr_type& cb_type) {
+  /* Build unique X-direction connection block modules */
+  vtr::Point<size_t> cb_range = device_rr_gsb.get_gsb_range();
+
+  for (size_t ix = 0; ix < cb_range.x(); ++ix) {
+    for (size_t iy = 0; iy < cb_range.y(); ++iy) {
+      /* Check if the connection block exists in the device!
+       * Some of them do NOT exist due to heterogeneous blocks (height > 1) 
+       * We will skip those modules
+       */
+      const RRGSB& rr_gsb = device_rr_gsb.get_gsb(ix, iy);
+      if (false == rr_gsb.is_cb_exist(cb_type)) {
+        continue;
+      }
+      print_pnr_sdc_constrain_cb_timing(sdc_dir,
+                                        module_manager,
+                                        rr_graph, 
+                                        rr_gsb, 
+                                        cb_type);
+
+    }
+  }
+}
+
+/********************************************************************
+ * Iterate over all the connection blocks in a device
+ * and print SDC file for each of them 
+ *******************************************************************/
+void print_pnr_sdc_flatten_routing_constrain_cb_timing(const std::string& sdc_dir,
+                                                       const ModuleManager& module_manager, 
+                                                       const RRGraph& rr_graph,
+                                                       const DeviceRRGSB& device_rr_gsb) {
+
+  /* Start time count */
+  vtr::ScopedStartFinishTimer timer("Write SDC for constrain Connection Block timing for P&R flow");
+
+  print_pnr_sdc_flatten_routing_constrain_cb_timing(sdc_dir, module_manager, 
+                                                    rr_graph,
+                                                    device_rr_gsb,
+                                                    CHANX);
+
+  print_pnr_sdc_flatten_routing_constrain_cb_timing(sdc_dir, module_manager, 
+                                                    rr_graph,
+                                                    device_rr_gsb,
+                                                    CHANY);
+}
+
+/********************************************************************
+ * Print SDC timing constraints for Connection blocks
+ * This function is designed for compact routing hierarchy
+ *******************************************************************/
+void print_pnr_sdc_compact_routing_constrain_cb_timing(const std::string& sdc_dir,
+                                                       const ModuleManager& module_manager,
+                                                       const RRGraph& rr_graph,
+                                                       const DeviceRRGSB& device_rr_gsb) {
+
+  /* Start time count */
+  vtr::ScopedStartFinishTimer timer("Write SDC for constrain Connection Block timing for P&R flow");
+
+  /* Print SDC for unique X-direction connection block modules */
+  for (size_t icb = 0; icb < device_rr_gsb.get_num_cb_unique_module(CHANX); ++icb) {
+    const RRGSB& unique_mirror = device_rr_gsb.get_cb_unique_module(CHANX, icb);
+    print_pnr_sdc_constrain_cb_timing(sdc_dir,
+                                      module_manager,
+                                      rr_graph, 
+                                      unique_mirror, 
+                                      CHANX);
+  }
+
+  /* Print SDC for unique Y-direction connection block modules */
+  for (size_t icb = 0; icb < device_rr_gsb.get_num_cb_unique_module(CHANY); ++icb) {
+    const RRGSB& unique_mirror = device_rr_gsb.get_cb_unique_module(CHANY, icb);
+    print_pnr_sdc_constrain_cb_timing(sdc_dir,
+                                      module_manager,
+                                      rr_graph, 
+                                      unique_mirror, 
+                                      CHANY);
+  }
+}
+
+} /* end namespace openfpga */

openfpga/src/fpga_sdc/pnr_sdc_routing_writer.h

@@ -0,0 +1,42 @@
+#ifndef PNR_SDC_ROUTING_WRITER_H
+#define PNR_SDC_ROUTING_WRITER_H
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+#include <string>
+#include <vector>
+#include "module_manager.h"
+#include "device_rr_gsb.h"
+#include "rr_graph_obj.h"
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+void print_pnr_sdc_flatten_routing_constrain_sb_timing(const std::string& sdc_dir,
+                                                       const ModuleManager& module_manager,
+                                                       const RRGraph& rr_graph,
+                                                       const DeviceRRGSB& device_rr_gsb);
+
+void print_pnr_sdc_compact_routing_constrain_sb_timing(const std::string& sdc_dir,
+                                                       const ModuleManager& module_manager,
+                                                       const RRGraph& rr_graph,
+                                                       const DeviceRRGSB& device_rr_gsb);
+
+void print_pnr_sdc_flatten_routing_constrain_cb_timing(const std::string& sdc_dir,
+                                                       const ModuleManager& module_manager, 
+                                                       const RRGraph& rr_graph,
+                                                       const DeviceRRGSB& device_rr_gsb);
+
+void print_pnr_sdc_compact_routing_constrain_cb_timing(const std::string& sdc_dir,
+                                                       const ModuleManager& module_manager,
+                                                       const RRGraph& rr_graph,
+                                                       const DeviceRRGSB& device_rr_gsb);
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/fpga_sdc/pnr_sdc_writer.cpp

@@ -0,0 +1,431 @@
+/********************************************************************
+ * This file includes functions that print SDC (Synopsys Design Constraint) 
+ * files in physical design tools, i.e., Place & Route (PnR) tools
+ * The SDC files are used to constrain the physical design for each module
+ * in FPGA fabric, such as Configurable Logic Blocks (CLBs), 
+ * Heterogeneous blocks, Switch Blocks (SBs) and Connection Blocks (CBs)
+ *
+ * Note that this is different from the SDC to constrain VPR Place&Route
+ * engine! These SDCs are designed for PnR to generate FPGA layouts!!!
+ *******************************************************************/
+#include <ctime>
+#include <fstream>
+#include <iomanip>
+
+/* Headers from vtrutil library */
+#include "vtr_assert.h"
+#include "vtr_time.h"
+#include "vtr_log.h"
+
+/* Headers from openfpgautil library */
+#include "openfpga_port.h"
+#include "openfpga_digest.h"
+
+#include "mux_utils.h"
+
+#include "openfpga_naming.h"
+
+#include "sdc_writer_naming.h"
+#include "sdc_writer_utils.h"
+#include "sdc_memory_utils.h"
+#include "pnr_sdc_routing_writer.h"
+#include "pnr_sdc_grid_writer.h"
+#include "pnr_sdc_writer.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * Print a SDC file to constrain the global ports of FPGA fabric
+ * in particular clock ports
+ * 
+ * For programming clock, we give a fixed period, while for operating
+ * clock, we constrain with critical path delay 
+ *******************************************************************/
+static 
+void print_pnr_sdc_global_ports(const std::string& sdc_dir, 
+                                const float& programming_critical_path_delay,
+                                const float& operating_critical_path_delay,
+                                const CircuitLibrary& circuit_lib,
+                                const std::vector<CircuitPortId>& global_ports) {
+
+  /* Create the file name for Verilog netlist */
+  std::string sdc_fname(sdc_dir + std::string(SDC_GLOBAL_PORTS_FILE_NAME));
+
+  /* Start time count */
+  std::string timer_message = std::string("Write SDC for constraining clocks for P&R flow '") + sdc_fname + std::string("'");
+  vtr::ScopedStartFinishTimer timer(timer_message);
+
+  /* Create the file stream */
+  std::fstream fp;
+  fp.open(sdc_fname, std::fstream::out | std::fstream::trunc);
+
+  check_file_stream(sdc_fname.c_str(), fp);
+
+  /* Generate the descriptions*/
+  print_sdc_file_header(fp, std::string("Clock contraints for PnR"));
+
+  /* Get clock port from the global port */
+  for (const CircuitPortId& clock_port : global_ports) {
+    if (CIRCUIT_MODEL_PORT_CLOCK != circuit_lib.port_type(clock_port)) {
+      continue;
+    }
+    /* Reach here, it means a clock port and we need print constraints */
+    float clock_period = operating_critical_path_delay; 
+
+    /* For programming clock, we give a fixed period */
+    if (true == circuit_lib.port_is_prog(clock_port)) {
+      clock_period = programming_critical_path_delay;
+      /* Print comments */
+      fp << "##################################################" << std::endl; 
+      fp << "# Create programmable clock                       " << std::endl;
+      fp << "##################################################" << std::endl; 
+    } else {
+      /* Print comments */
+      fp << "##################################################" << std::endl; 
+      fp << "# Create clock                                    " << std::endl;
+      fp << "##################################################" << std::endl; 
+    }
+
+    for (const size_t& pin : circuit_lib.pins(clock_port)) {
+      BasicPort port_to_constrain(circuit_lib.port_prefix(clock_port), pin, pin);
+
+      fp << "create_clock ";
+      fp << generate_sdc_port(port_to_constrain) << "-period ";
+      fp << std::setprecision(10) << clock_period;
+      fp << " -waveform {0 ";
+      fp << std::setprecision(10) << clock_period / 2;
+      fp << "}" << std::endl;
+
+      fp << std::endl;
+    }
+  }
+
+  /* For non-clock port from the global port: give a fixed period */
+  for (const CircuitPortId& global_port : global_ports) {
+    if (CIRCUIT_MODEL_PORT_CLOCK == circuit_lib.port_type(global_port)) {
+      continue;
+    }
+
+    /* Print comments */
+    fp << "##################################################" << std::endl; 
+    fp << "# Constrain other global ports                    " << std::endl;
+    fp << "##################################################" << std::endl; 
+
+    /* Reach here, it means a non-clock global port and we need print constraints */
+    float clock_period = operating_critical_path_delay; 
+    for (const size_t& pin : circuit_lib.pins(global_port)) {
+      BasicPort port_to_constrain(circuit_lib.port_prefix(global_port), pin, pin);
+      fp << "create_clock ";
+      fp << generate_sdc_port(port_to_constrain) << "-period ";
+      fp << std::setprecision(10) << clock_period;
+      fp << " -waveform {0 ";
+      fp << std::setprecision(10) << clock_period / 2;
+      fp << "} ";
+      fp << "[list [get_ports { " << generate_sdc_port(port_to_constrain) << "}]]" << std::endl;
+
+      fp << "set_drive 0 " << generate_sdc_port(port_to_constrain) << std::endl;
+     
+      fp << std::endl;
+    }
+  }
+
+  /* Close file handler */
+  fp.close();
+}
+
+/********************************************************************
+ * Break combinational loops in FPGA fabric, which mainly come from
+ * configurable memory cells. 
+ * To handle this, we disable the outputs of memory cells
+ *******************************************************************/
+static 
+void print_pnr_sdc_constrain_configurable_memory_outputs(const std::string& sdc_dir,
+                                                         const ModuleManager& module_manager,
+                                                         const ModuleId& top_module) {
+
+  /* Create the file name for Verilog netlist */
+  std::string sdc_fname(sdc_dir + std::string(SDC_DISABLE_CONFIG_MEM_OUTPUTS_FILE_NAME));
+
+  /* Start time count */
+  std::string timer_message = std::string("Write SDC to disable configurable memory outputs for P&R flow '") + sdc_fname + std::string("'");
+  vtr::ScopedStartFinishTimer timer(timer_message);
+
+  /* Create the file stream */
+  std::fstream fp;
+  fp.open(sdc_fname, std::fstream::out | std::fstream::trunc);
+
+  check_file_stream(sdc_fname.c_str(), fp);
+
+  /* Generate the descriptions*/
+  print_sdc_file_header(fp, std::string("Disable configurable memory outputs for PnR"));
+
+  /* Go recursively in the module manager, starting from the top-level module: instance id of the top-level module is 0 by default */
+  rec_print_pnr_sdc_disable_configurable_memory_module_output(fp, module_manager, top_module, 
+                                                              format_dir_path(module_manager.module_name(top_module)));
+
+  /* Close file handler */
+  fp.close();
+}
+
+/********************************************************************
+ * Break combinational loops in FPGA fabric, which mainly come from 
+ * loops of multiplexers.
+ * To handle this, we disable the timing at outputs of routing multiplexers
+ *******************************************************************/
+static 
+void print_sdc_disable_routing_multiplexer_outputs(const std::string& sdc_dir,
+                                                   const MuxLibrary& mux_lib,
+                                                   const CircuitLibrary& circuit_lib,
+                                                   const ModuleManager& module_manager) {
+  /* Create the file name for Verilog netlist */
+  std::string sdc_fname(sdc_dir + std::string(SDC_DISABLE_MUX_OUTPUTS_FILE_NAME));
+
+  /* Start time count */
+  std::string timer_message = std::string("Write SDC to disable routing multiplexer outputs for P&R flow '") + sdc_fname + std::string("'");
+  vtr::ScopedStartFinishTimer timer(timer_message);
+
+  /* Create the file stream */
+  std::fstream fp;
+  fp.open(sdc_fname, std::fstream::out | std::fstream::trunc);
+
+  check_file_stream(sdc_fname.c_str(), fp);
+
+  /* Generate the descriptions*/
+  print_sdc_file_header(fp, std::string("Disable routing multiplexer outputs for PnR"));
+
+  /* Iterate over the MUX modules */
+  for (const MuxId& mux_id : mux_lib.muxes()) {
+    const CircuitModelId& mux_model = mux_lib.mux_circuit_model(mux_id);
+    
+    /* Skip LUTs, we only care about multiplexers here */
+    if (CIRCUIT_MODEL_MUX != circuit_lib.model_type(mux_model)) {
+      continue;
+    }
+
+    const MuxGraph& mux_graph = mux_lib.mux_graph(mux_id);
+    std::string mux_module_name = generate_mux_subckt_name(circuit_lib, mux_model, 
+                                                           find_mux_num_datapath_inputs(circuit_lib, mux_model, mux_graph.num_inputs()), 
+                                                           std::string(""));
+    /* Find the module name in module manager */
+    ModuleId mux_module = module_manager.find_module(mux_module_name);
+    VTR_ASSERT(true == module_manager.valid_module_id(mux_module));
+     
+    /* Disable the timing for the output ports */ 
+    for (const BasicPort& output_port : module_manager.module_ports_by_type(mux_module, ModuleManager::MODULE_OUTPUT_PORT)) {
+      fp << "set_disable_timing [get_pins -filter \"name =~ " << output_port.get_name() << "*\" ";
+      fp << "-of [get_cells -hier -filter \"ref_lib_cell_name == " << mux_module_name << "\"]]" << std::endl;
+      fp << std::endl;
+    }
+  }
+
+  /* Close file handler */
+  fp.close();
+}
+
+/********************************************************************
+ * Break combinational loops in FPGA fabric, which mainly come from 
+ * loops of multiplexers.
+ * To handle this, we disable the timing at outputs of Switch blocks
+ * This function is designed for flatten routing hierarchy
+ *******************************************************************/
+static 
+void print_pnr_sdc_flatten_routing_disable_switch_block_outputs(const std::string& sdc_dir,
+                                                                const ModuleManager& module_manager,
+                                                                const DeviceRRGSB& device_rr_gsb) {
+  /* Create the file name for Verilog netlist */
+  std::string sdc_fname(sdc_dir + std::string(SDC_DISABLE_SB_OUTPUTS_FILE_NAME));
+
+  /* Start time count */
+  std::string timer_message = std::string("Write SDC to disable switch block outputs for P&R flow '") + sdc_fname + std::string("'");
+  vtr::ScopedStartFinishTimer timer(timer_message);
+
+  /* Create the file stream */
+  std::fstream fp;
+  fp.open(sdc_fname, std::fstream::out | std::fstream::trunc);
+
+  check_file_stream(sdc_fname.c_str(), fp);
+
+  /* Generate the descriptions*/
+  print_sdc_file_header(fp, std::string("Disable Switch Block outputs for PnR"));
+
+  /* Get the range of SB array */
+  vtr::Point<size_t> sb_range = device_rr_gsb.get_gsb_range();
+  /* Go for each SB */
+  for (size_t ix = 0; ix < sb_range.x(); ++ix) {
+    for (size_t iy = 0; iy < sb_range.y(); ++iy) {
+      const RRGSB& rr_gsb = device_rr_gsb.get_gsb(ix, iy);
+
+      if (false == rr_gsb.is_sb_exist()) {
+        continue;
+      }
+
+      vtr::Point<size_t> gsb_coordinate(rr_gsb.get_sb_x(), rr_gsb.get_sb_y());
+      std::string sb_instance_name = generate_switch_block_module_name(gsb_coordinate); 
+
+      ModuleId sb_module = module_manager.find_module(sb_instance_name);
+      VTR_ASSERT(true == module_manager.valid_module_id(sb_module));
+
+      /* Disable the outputs of the module */
+      for (const BasicPort& output_port : module_manager.module_ports_by_type(sb_module, ModuleManager::MODULE_OUTPUT_PORT)) {
+        fp << "set_disable_timing " << sb_instance_name << "/" << output_port.get_name() << std::endl;
+        fp << std::endl;
+      }
+    }
+  }
+  
+  /* Close file handler */
+  fp.close();
+}
+
+/********************************************************************
+ * Break combinational loops in FPGA fabric, which mainly come from 
+ * loops of multiplexers.
+ * To handle this, we disable the timing at outputs of Switch blocks
+ * This function is designed for compact routing hierarchy
+ *******************************************************************/
+static 
+void print_pnr_sdc_compact_routing_disable_switch_block_outputs(const std::string& sdc_dir,
+                                                                const ModuleManager& module_manager,
+                                                                const ModuleId& top_module,
+                                                                const DeviceRRGSB& device_rr_gsb) {
+  /* Create the file name for Verilog netlist */
+  std::string sdc_fname(sdc_dir + std::string(SDC_DISABLE_SB_OUTPUTS_FILE_NAME));
+
+  /* Start time count */
+  std::string timer_message = std::string("Write SDC to disable switch block outputs for P&R flow '") + sdc_fname + std::string("'");
+  vtr::ScopedStartFinishTimer timer(timer_message);
+
+  /* Create the file stream */
+  std::fstream fp;
+  fp.open(sdc_fname, std::fstream::out | std::fstream::trunc);
+
+  check_file_stream(sdc_fname.c_str(), fp);
+
+  /* Generate the descriptions*/
+  print_sdc_file_header(fp, std::string("Disable Switch Block outputs for PnR"));
+
+  /* Build unique switch block modules */
+  for (size_t isb = 0; isb < device_rr_gsb.get_num_sb_unique_module(); ++isb) {
+    const RRGSB& rr_gsb = device_rr_gsb.get_sb_unique_module(isb);
+    vtr::Point<size_t> gsb_coordinate(rr_gsb.get_sb_x(), rr_gsb.get_sb_y());
+    std::string sb_module_name = generate_switch_block_module_name(gsb_coordinate); 
+
+    ModuleId sb_module = module_manager.find_module(sb_module_name);
+    VTR_ASSERT(true == module_manager.valid_module_id(sb_module));
+
+    /* Find all the instances in the top-level module */
+    for (const size_t& instance_id : module_manager.child_module_instances(top_module, sb_module)) {
+      std::string sb_instance_name = module_manager.instance_name(top_module, sb_module, instance_id);
+      /* Disable the outputs of the module */
+      for (const BasicPort& output_port : module_manager.module_ports_by_type(sb_module, ModuleManager::MODULE_OUTPUT_PORT)) {
+        fp << "set_disable_timing " << sb_instance_name << "/" << output_port.get_name() << std::endl;
+        fp << std::endl;
+      }
+    }
+  }
+  
+  /* Close file handler */
+  fp.close();
+}
+
+/********************************************************************
+ * Top-level function to print a number of SDC files in different purpose
+ * This function will generate files upon the options provided by users
+ * 1. Design constraints for CLBs
+ * 2. Design constraints for Switch Blocks
+ * 3. Design constraints for Connection Blocks 
+ * 4. Design constraints for breaking the combinational loops in FPGA fabric
+ *******************************************************************/
+void print_pnr_sdc(const PnrSdcOption& sdc_options,
+                   const float& programming_critical_path_delay,
+                   const float& operating_critical_path_delay,
+                   const DeviceContext& device_ctx,
+                   const VprDeviceAnnotation& device_annotation,
+                   const DeviceRRGSB& device_rr_gsb,
+                   const ModuleManager& module_manager,
+                   const MuxLibrary& mux_lib,
+                   const CircuitLibrary& circuit_lib,
+                   const std::vector<CircuitPortId>& global_ports,
+                   const bool& compact_routing_hierarchy) {
+  
+  /* Constrain global ports */
+  if (true == sdc_options.constrain_global_port()) {
+    print_pnr_sdc_global_ports(sdc_options.sdc_dir(),
+                               programming_critical_path_delay,
+                               operating_critical_path_delay,
+                               circuit_lib, global_ports);
+  }
+
+  std::string top_module_name = generate_fpga_top_module_name();
+  ModuleId top_module = module_manager.find_module(top_module_name);
+  VTR_ASSERT(true == module_manager.valid_module_id(top_module));
+
+  /* Output Design Constraints to disable outputs of memory cells */
+  if (true == sdc_options.constrain_configurable_memory_outputs()) {
+    print_pnr_sdc_constrain_configurable_memory_outputs(sdc_options.sdc_dir(), module_manager, top_module); 
+  } 
+
+  /* Break loops from Multiplexer Output */
+  if (true == sdc_options.constrain_routing_multiplexer_outputs()) {
+    print_sdc_disable_routing_multiplexer_outputs(sdc_options.sdc_dir(),
+                                                  mux_lib, circuit_lib,
+                                                  module_manager);
+  }
+
+  /* Break loops from any SB output */
+  if (true == sdc_options.constrain_switch_block_outputs()) {
+    if (true == compact_routing_hierarchy) {
+      print_pnr_sdc_compact_routing_disable_switch_block_outputs(sdc_options.sdc_dir(),
+                                                                 module_manager, top_module,
+                                                                 device_rr_gsb);
+    } else {
+      VTR_ASSERT_SAFE (false == compact_routing_hierarchy);
+      print_pnr_sdc_flatten_routing_disable_switch_block_outputs(sdc_options.sdc_dir(),
+                                                                 module_manager,
+                                                                 device_rr_gsb);
+    }
+  }
+
+  /* Output routing constraints for Switch Blocks */
+  if (true == sdc_options.constrain_sb()) {
+    if (true == compact_routing_hierarchy) {
+      print_pnr_sdc_compact_routing_constrain_sb_timing(sdc_options.sdc_dir(),
+                                                        module_manager,
+                                                        device_ctx.rr_graph,
+                                                        device_rr_gsb);
+    } else {
+	  VTR_ASSERT_SAFE (false == compact_routing_hierarchy);
+      print_pnr_sdc_flatten_routing_constrain_sb_timing(sdc_options.sdc_dir(),
+                                                        module_manager,
+                                                        device_ctx.rr_graph,
+                                                        device_rr_gsb);
+    }
+  }
+
+  /* Output routing constraints for Connection Blocks */
+  if (true == sdc_options.constrain_cb()) {
+    if (true == compact_routing_hierarchy) {
+      print_pnr_sdc_compact_routing_constrain_cb_timing(sdc_options.sdc_dir(),
+                                                        module_manager,
+                                                        device_ctx.rr_graph,
+                                                        device_rr_gsb);
+    } else {
+	  VTR_ASSERT_SAFE (false == compact_routing_hierarchy);
+      print_pnr_sdc_flatten_routing_constrain_cb_timing(sdc_options.sdc_dir(),
+                                                        module_manager, 
+                                                        device_ctx.rr_graph,
+                                                        device_rr_gsb);
+    }
+  }
+
+  /* Output Timing constraints for Programmable blocks */
+  if (true == sdc_options.constrain_grid()) {
+    print_pnr_sdc_constrain_grid_timing(sdc_options.sdc_dir(),
+                                        device_ctx,
+                                        device_annotation,
+                                        module_manager);
+  }
+}
+
+} /* end namespace openfpga */

openfpga/src/fpga_sdc/pnr_sdc_writer.h

@@ -0,0 +1,38 @@
+#ifndef PNR_SDC_WRITER_H
+#define PNR_SDC_WRITER_H
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+#include <string>
+#include <vector>
+#include "vpr_context.h"
+#include "vpr_device_annotation.h"
+#include "device_rr_gsb.h"
+#include "module_manager.h"
+#include "mux_library.h"
+#include "circuit_library.h"
+#include "pnr_sdc_option.h"
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+void print_pnr_sdc(const PnrSdcOption& sdc_options,
+                   const float& programming_critical_path_delay,
+                   const float& operating_critical_path_delay,
+                   const DeviceContext& device_ctx,
+                   const VprDeviceAnnotation& device_annotation,
+                   const DeviceRRGSB& device_rr_gsb,
+                   const ModuleManager& module_manager,
+                   const MuxLibrary& mux_lib,
+                   const CircuitLibrary& circuit_lib,
+                   const std::vector<CircuitPortId>& global_ports,
+                   const bool& compact_routing_hierarchy);
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/fpga_sdc/sdc_memory_utils.cpp

@@ -0,0 +1,69 @@
+/********************************************************************
+ * Most utilized function used to constrain memory cells in FPGA
+ * fabric using SDC commands
+ *******************************************************************/
+
+/* Headers from openfpgautil library */
+#include "openfpga_digest.h"
+
+#include "sdc_writer_utils.h"
+
+#include "sdc_memory_utils.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * Print SDC commands to disable outputs of all the configurable memory modules
+ * in a given module 
+ * This function will be executed in a recursive way, 
+ * using a Depth-First Search (DFS) strategy
+ * It will iterate over all the configurable children under each module
+ * and print a SDC command to disable its outputs
+ *******************************************************************/
+void rec_print_pnr_sdc_disable_configurable_memory_module_output(std::fstream& fp, 
+                                                                 const ModuleManager& module_manager, 
+                                                                 const ModuleId& parent_module,
+                                                                 const std::string& parent_module_path) {
+
+  /* For each configurable child, we will go one level down in priority */
+  for (size_t child_index = 0; child_index < module_manager.configurable_children(parent_module).size(); ++child_index) {
+    std::string child_module_path = parent_module_path;
+    ModuleId child_module_id = module_manager.configurable_children(parent_module)[child_index];
+    size_t child_instance_id = module_manager.configurable_child_instances(parent_module)[child_index];
+    if (true == module_manager.instance_name(parent_module, child_module_id, child_instance_id).empty()) {
+      /* Give a default name <module_name>_<instance_id>_ */
+      child_module_path += module_manager.module_name(child_module_id); 
+      child_module_path += "_";
+      child_module_path += std::to_string(child_instance_id);
+      child_module_path += "_";
+    } else {
+      child_module_path += module_manager.instance_name(parent_module, child_module_id, child_instance_id);
+    }
+    child_module_path = format_dir_path(child_module_path);
+
+    rec_print_pnr_sdc_disable_configurable_memory_module_output(fp, module_manager, 
+                                                                child_module_id, 
+                                                                child_module_path);
+  }
+
+  /* If there is no configurable children any more, this is a leaf module, print a SDC command for disable timing */
+  if (0 < module_manager.configurable_children(parent_module).size()) {
+    return;
+  }
+
+  /* Validate file stream */
+  valid_file_stream(fp);
+
+  /* Disable timing for each output port of this module */
+  for (const BasicPort& output_port : module_manager.module_ports_by_type(parent_module, ModuleManager::MODULE_OUTPUT_PORT)) {
+    for (const size_t& pin : output_port.pins()) {
+      BasicPort output_pin(output_port.get_name(), pin, pin);
+      fp << "set_disable_timing ";
+      fp << parent_module_path << generate_sdc_port(output_pin);
+      fp << std::endl;
+    }
+  }
+}
+
+} /* end namespace openfpga */

openfpga/src/fpga_sdc/sdc_memory_utils.h

@@ -0,0 +1,25 @@
+#ifndef SDC_MEMORY_UTILS_H
+#define SDC_MEMORY_UTILS_H
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+#include <fstream>
+#include <string>
+#include "module_manager.h"
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+void rec_print_pnr_sdc_disable_configurable_memory_module_output(std::fstream& fp, 
+                                                                 const ModuleManager& module_manager, 
+                                                                 const ModuleId& parent_module,
+                                                                 const std::string& parent_module_path);
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/fpga_sdc/sdc_writer_naming.h

@@ -0,0 +1,20 @@
+#ifndef SDC_WRITER_NAMING_H
+#define SDC_WRITER_NAMING_H
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+constexpr char* SDC_FILE_NAME_POSTFIX = ".sdc";
+
+constexpr char* SDC_GLOBAL_PORTS_FILE_NAME = "global_ports.sdc";
+constexpr char* SDC_BENCHMARK_ANALYSIS_FILE_NAME= "fpga_top_analysis.sdc";
+constexpr char* SDC_DISABLE_CONFIG_MEM_OUTPUTS_FILE_NAME = "disable_configurable_memory_outputs.sdc";
+constexpr char* SDC_DISABLE_MUX_OUTPUTS_FILE_NAME = "disable_routing_multiplexer_outputs.sdc";
+constexpr char* SDC_DISABLE_SB_OUTPUTS_FILE_NAME = "disable_sb_outputs.sdc";
+constexpr char* SDC_CB_FILE_NAME = "cb.sdc";
+
+constexpr char* SDC_ANALYSIS_FILE_NAME = "fpga_top_analysis.sdc";
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/fpga_sdc/sdc_writer_utils.cpp

@@ -0,0 +1,202 @@
+/********************************************************************
+ * This file include most utilized functions to be used in SDC writers 
+ *******************************************************************/
+#include <chrono>
+#include <ctime>
+#include <iomanip>
+
+/* Headers from openfpgautil library */
+#include "openfpga_digest.h"
+
+#include "sdc_writer_utils.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * Write a head (description) in SDC file 
+ *******************************************************************/
+void print_sdc_file_header(std::fstream& fp,
+                           const std::string& usage) {
+
+  valid_file_stream(fp);
+
+  auto end = std::chrono::system_clock::now(); 
+  std::time_t end_time = std::chrono::system_clock::to_time_t(end);
+
+  fp << "#############################################" << std::endl;
+  fp << "#\tSynopsys Design Constraints (SDC)" << std::endl;
+  fp << "#\tFor FPGA fabric " << std::endl;
+  fp << "#\tDescription: " << usage << std::endl;
+  fp << "#\tAuthor: Xifan TANG " << std::endl;
+  fp << "#\tOrganization: University of Utah " << std::endl;
+  fp << "#\tDate: " << std::ctime(&end_time);
+  fp << "#############################################" << std::endl;
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Write a port in SDC format
+ *******************************************************************/
+std::string generate_sdc_port(const BasicPort& port) {
+  std::string sdc_line;
+
+  std::string size_str = "[" + std::to_string(port.get_lsb()) + ":" + std::to_string(port.get_msb()) + "]";
+
+  /* Only connection require a format of <port_name>[<lsb>:<msb>]
+   * others require a format of <port_type> [<lsb>:<msb>] <port_name> 
+   */
+  /* When LSB == MSB, we can use a simplified format <port_type>[<lsb>]*/
+  if ( 1 == port.get_width()) {
+    size_str = "[" + std::to_string(port.get_lsb()) + "]";
+  }
+  sdc_line = port.get_name() + size_str;
+
+  return sdc_line;
+}
+
+/********************************************************************
+ * Constrain a path between two ports of a module with a given maximum timing value
+ *******************************************************************/
+void print_pnr_sdc_constrain_max_delay(std::fstream& fp,
+                                       const std::string& src_instance_name,
+                                       const std::string& src_port_name,
+                                       const std::string& des_instance_name,
+                                       const std::string& des_port_name,
+                                       const float& delay) {
+  /* Validate file stream */
+  valid_file_stream(fp);
+
+  fp << "set_max_delay";
+
+  fp << " -from ";
+  if (!src_instance_name.empty()) {
+    fp << src_instance_name << "/";
+  }
+  fp << src_port_name;
+
+  fp << " -to ";
+ 
+  if (!des_instance_name.empty()) {
+    fp << des_instance_name << "/";
+  }
+  fp << des_port_name;
+
+  fp << " " << std::setprecision(10) << delay;
+
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Constrain a path between two ports of a module with a given timing value
+ * Note: this function uses set_max_delay !!!
+ *******************************************************************/
+void print_pnr_sdc_constrain_module_port2port_timing(std::fstream& fp,
+                                                     const ModuleManager& module_manager,
+                                                     const ModuleId& input_parent_module_id, 
+                                                     const ModulePortId& module_input_port_id, 
+                                                     const ModuleId& output_parent_module_id, 
+                                                     const ModulePortId& module_output_port_id, 
+                                                     const float& tmax) {
+  print_pnr_sdc_constrain_max_delay(fp,
+                                    module_manager.module_name(input_parent_module_id),
+                                    generate_sdc_port(module_manager.module_port(input_parent_module_id, module_input_port_id)),
+                                    module_manager.module_name(output_parent_module_id),
+                                    generate_sdc_port(module_manager.module_port(output_parent_module_id, module_output_port_id)),
+                                    tmax);
+
+}
+
+/********************************************************************
+ * Constrain a path between two ports of a module with a given timing value
+ * This function will NOT output the module name
+ * Note: this function uses set_max_delay !!!
+ *******************************************************************/
+void print_pnr_sdc_constrain_port2port_timing(std::fstream& fp,
+                                              const ModuleManager& module_manager,
+                                              const ModuleId& input_parent_module_id, 
+                                              const ModulePortId& module_input_port_id, 
+                                              const ModuleId& output_parent_module_id, 
+                                              const ModulePortId& module_output_port_id, 
+                                              const float& tmax) {
+  print_pnr_sdc_constrain_max_delay(fp,
+                                    std::string(),
+                                    generate_sdc_port(module_manager.module_port(input_parent_module_id, module_input_port_id)),
+                                    std::string(),
+                                    generate_sdc_port(module_manager.module_port(output_parent_module_id, module_output_port_id)),
+                                    tmax);
+
+}
+
+/********************************************************************
+ * Disable timing for a port 
+ *******************************************************************/
+void print_sdc_disable_port_timing(std::fstream& fp,
+                                   const BasicPort& port) {
+  /* Validate file stream */
+  valid_file_stream(fp);
+
+  fp << "set_disable_timing ";
+
+  fp << generate_sdc_port(port);
+
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Set the input delay for a port in SDC format 
+ * Note that the input delay will be bounded by a clock port
+ *******************************************************************/
+void print_sdc_set_port_input_delay(std::fstream& fp,
+                                    const BasicPort& port,
+                                    const BasicPort& clock_port,
+                                    const float& delay) {
+  /* Validate file stream */
+  valid_file_stream(fp);
+
+  fp << "set_input_delay ";
+
+  fp << "-clock ";
+
+  fp << generate_sdc_port(clock_port);
+
+  fp << " -max ";
+ 
+  fp << std::setprecision(10) << delay;
+
+  fp << " ";
+
+  fp << generate_sdc_port(port);
+
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Set the output delay for a port in SDC format 
+ * Note that the output delay will be bounded by a clock port
+ *******************************************************************/
+void print_sdc_set_port_output_delay(std::fstream& fp,
+                                     const BasicPort& port,
+                                     const BasicPort& clock_port,
+                                     const float& delay) {
+  /* Validate file stream */
+  valid_file_stream(fp);
+
+  fp << "set_output_delay ";
+
+  fp << "-clock ";
+
+  fp << generate_sdc_port(clock_port);
+
+  fp << " -max ";
+ 
+  fp << std::setprecision(10) << delay;
+
+  fp << " ";
+
+  fp << generate_sdc_port(port);
+
+  fp << std::endl;
+}
+
+} /* end namespace openfpga */

openfpga/src/fpga_sdc/sdc_writer_utils.h

@@ -0,0 +1,62 @@
+#ifndef SDC_WRITER_UTILS_H
+#define SDC_WRITER_UTILS_H
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+#include <fstream>
+#include <string>
+#include "openfpga_port.h"
+#include "module_manager.h"
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+void print_sdc_file_header(std::fstream& fp,
+                           const std::string& usage);
+
+std::string generate_sdc_port(const BasicPort& port);
+
+void print_pnr_sdc_constrain_max_delay(std::fstream& fp,
+                                       const std::string& src_instance_name,
+                                       const std::string& src_port_name,
+                                       const std::string& des_instance_name,
+                                       const std::string& des_port_name,
+                                       const float& delay);
+
+void print_pnr_sdc_constrain_module_port2port_timing(std::fstream& fp,
+                                                     const ModuleManager& module_manager,
+                                                     const ModuleId& input_parent_module_id, 
+                                                     const ModulePortId& module_input_port_id, 
+                                                     const ModuleId& output_parent_module_id, 
+                                                     const ModulePortId& module_output_port_id, 
+                                                     const float& tmax);
+
+void print_pnr_sdc_constrain_port2port_timing(std::fstream& fp,
+                                              const ModuleManager& module_manager,
+                                              const ModuleId& input_parent_module_id, 
+                                              const ModulePortId& module_input_port_id, 
+                                              const ModuleId& output_parent_module_id, 
+                                              const ModulePortId& module_output_port_id, 
+                                              const float& tmax);
+
+void print_sdc_disable_port_timing(std::fstream& fp,
+                                   const BasicPort& port);
+
+void print_sdc_set_port_input_delay(std::fstream& fp,
+                                    const BasicPort& port,
+                                    const BasicPort& clock_port,
+                                    const float& delay);
+
+void print_sdc_set_port_output_delay(std::fstream& fp,
+                                     const BasicPort& port,
+                                     const BasicPort& clock_port,
+                                     const float& delay);
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/fpga_verilog/fabric_verilog_options.cpp

@@ -3,7 +3,7 @@
  ******************************************************************************/
 #include "vtr_assert.h"
 
-#include "verilog_options.h"
+#include "fabric_verilog_options.h"
 
 /* begin namespace openfpga */
 namespace openfpga {
@@ -52,22 +52,6 @@ bool FabricVerilogOption::compress_routing() const {
   return compress_routing_;
 }
 
-bool FabricVerilogOption::print_top_testbench() const {
-  return print_top_testbench_;
-}
-
-bool FabricVerilogOption::print_formal_verification_top_netlist() const {
-  return print_formal_verification_top_netlist_;
-}
-
-bool FabricVerilogOption::print_autocheck_top_testbench() const {
-  return false == reference_verilog_file_path_.empty();
-}
-
-std::string FabricVerilogOption::reference_verilog_file_path() const {
-  return reference_verilog_file_path_;
-}
-
 bool FabricVerilogOption::print_user_defined_template() const {
   return print_user_defined_template_;
 }
@@ -103,18 +87,6 @@ void FabricVerilogOption::set_compress_routing(const bool& enabled) {
   compress_routing_ = enabled;
 }
 
-void FabricVerilogOption::set_print_top_testbench(const bool& enabled) {
-  print_top_testbench_ = enabled;
-}
-
-void FabricVerilogOption::set_print_formal_verification_top_netlist(const bool& enabled) {
-  print_formal_verification_top_netlist_ = enabled;
-}
-
-void FabricVerilogOption::set_print_autocheck_top_testbench(const std::string& reference_verilog_file_path) {
-  reference_verilog_file_path_ = reference_verilog_file_path;
-}
-
 void FabricVerilogOption::set_print_user_defined_template(const bool& enabled) {
   print_user_defined_template_ = enabled;
 }

openfpga/src/fpga_verilog/fabric_verilog_options.h

@@ -1,5 +1,5 @@
-#ifndef VERILOG_OPTIONS_H
+#ifndef FABRIC_VERILOG_OPTIONS_H
-#define VERILOG_OPTIONS_H
+#define FABRIC_VERILOG_OPTIONS_H
 
 /********************************************************************
  * Include header files required by the data structure definition
@@ -10,11 +10,7 @@
 namespace openfpga {
 
 /********************************************************************
- * FlowManager aims to resolve the dependency between OpenFPGA functional
+ * Options for Fabric Verilog generator
- * code blocks
- * It can provide flags for downstream modules about if the data structures
- * they require have already been constructed
- *
  *******************************************************************/
 class FabricVerilogOption {
   public: /* Public constructor */

openfpga/src/fpga_verilog/simulation_info_writer.cpp

@@ -0,0 +1,67 @@
+/*********************************************************************
+ * This function includes the writer for generating exchangeable
+ * information, in order to interface different simulators   
+ ********************************************************************/
+#include <cmath>
+#include <ctime>
+#include <map>
+#define MINI_CASE_SENSITIVE
+#include "ini.h"
+
+/* Headers from vtrutil library */
+#include "vtr_assert.h"
+#include "vtr_time.h"
+
+#include "simulation_utils.h"
+
+#include "verilog_constants.h"
+#include "simulation_info_writer.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/*********************************************************************
+ * Top-level function to write an ini file which contains exchangeable
+ * information, in order to interface different Verilog simulators
+ ********************************************************************/
+void print_verilog_simulation_info(const std::string& ini_fname,
+                                   const std::string& circuit_name,
+                                   const std::string& src_dir,
+                                   const size_t& num_program_clock_cycles,
+                                   const int& num_operating_clock_cycles,
+                                   const float& prog_clock_freq,
+                                   const float& op_clock_freq) {
+
+  std::string timer_message = std::string("Write exchangeable file containing simulation information '") + ini_fname + std::string("'");
+
+  /* Start time count */
+  vtr::ScopedStartFinishTimer timer(timer_message);
+
+  /* Use default name if user does not provide one */
+  VTR_ASSERT(true != ini_fname.empty());
+
+  mINI::INIStructure ini;
+  // std::map<char, int> units_map;
+  // units_map['s']=1;  // units_map['ms']=1E-3;  // units_map['us']=1E-6;
+  // units_map['ns']=1E-9;  // units_map['ps']=1E-12;  // units_map['fs']=1E-15;
+
+  /* Compute simulation time period */
+  float simulation_time_period = find_simulation_time_period(1E-3,
+                                                             num_program_clock_cycles,
+                                                             1. / prog_clock_freq,
+                                                             num_operating_clock_cycles,
+                                                             1. / op_clock_freq);
+  ini["SIMULATION_DECK"]["PROJECTNAME "] = "ModelSimProject";
+  ini["SIMULATION_DECK"]["BENCHMARK "] = circuit_name;
+  ini["SIMULATION_DECK"]["TOP_TB"] = circuit_name + std::string(FORMAL_RANDOM_TOP_TESTBENCH_POSTFIX);
+  ini["SIMULATION_DECK"]["SIMTIME "] = std::to_string(simulation_time_period);
+  ini["SIMULATION_DECK"]["UNIT "] = "ms";
+  ini["SIMULATION_DECK"]["VERILOG_PATH "] = std::string(src_dir);
+  ini["SIMULATION_DECK"]["VERILOG_FILE1"] = std::string(DEFINES_VERILOG_FILE_NAME);
+  ini["SIMULATION_DECK"]["VERILOG_FILE2"] = std::string(circuit_name + std::string(TOP_INCLUDE_NETLIST_FILE_NAME_POSTFIX));
+
+  mINI::INIFile file(ini_fname);
+  file.generate(ini, true);
+}
+
+} /* end namespace openfpga */

openfpga/src/fpga_verilog/simulation_info_writer.h

@@ -0,0 +1,26 @@
+#ifndef SIMULATION_INFO_WRITER_H
+#define SIMULATION_INFO_WRITER_H
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+#include <string>
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+void print_verilog_simulation_info(const std::string& ini_fname,
+                                   const std::string& circuit_name,
+                                   const std::string& src_dir,
+                                   const size_t& num_program_clock_cycles,
+                                   const int& num_operating_clock_cycles,
+                                   const float& prog_clock_freq,
+                                   const float& op_clock_freq);
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/fpga_verilog/verilog_api.cpp

@@ -20,6 +20,11 @@
 #include "verilog_grid.h"
 #include "verilog_top_module.h"
 
+#include "verilog_preconfig_top_module.h"
+#include "verilog_formal_random_top_testbench.h"
+#include "verilog_top_testbench.h"
+#include "simulation_info_writer.h"
+
 /* Header file for this source file */
 #include "verilog_api.h"
 
@@ -27,17 +32,18 @@
 namespace openfpga {
 
 /********************************************************************
- * Top-level function of FPGA-Verilog
+ * A top-level function of FPGA-Verilog which focuses on fabric Verilog generation
  * This function will generate
- * 1. primitive modules required by the full fabric
+ *  - primitive modules required by the full fabric
- *    which are LUTs, routing multiplexer, logic gates, transmission-gates etc.
+ *  - which are LUTs, routing multiplexer, logic gates, transmission-gates etc.
- * 2. Routing modules, which are Switch Blocks (SBs) and Connection Blocks (CBs)
+ *  - Routing modules, which are Switch Blocks (SBs) and Connection Blocks (CBs)
- * 3. Logic block modules, which are Configuration Logic Blocks (CLBs)
+ *  - Logic block modules, which are Configuration Logic Blocks (CLBs)
- * 4. FPGA module, which are the full FPGA fabric with configuration protocol
+ *  - FPGA module, which are the full FPGA fabric with configuration protocol
- * 5. A wrapper module, which encapsulate the FPGA module in a Verilog module which have the same port as the input benchmark 
+ *
- * 6. Testbench, where a FPGA module is configured with a bitstream and then driven by input vectors
+ * Note: 
- * 7. Pre-configured testbench, which can skip the configuration phase and pre-configure the FPGA module. This testbench is created for quick verification and formal verification purpose.
+ *  - Please do NOT include ANY testbench generation in this function!!!
- * 8. Verilog netlist including preprocessing flags and all the Verilog netlists that have been generated
+ *    It is about the fabric itself, independent from any implementation
+ *    All the testbench generation should be in the function fpga_testbench_verilog()
  *
  * TODO: We should use module manager as a constant here. 
  * All the modification should be done before this writer!
@@ -75,9 +81,6 @@ void fpga_fabric_verilog(ModuleManager& module_manager,
   print_verilog_preprocessing_flags_netlist(std::string(src_dir_path),
                                             options);
 
-  print_verilog_simulation_preprocessing_flags(std::string(src_dir_path),
-                                               options);
-
   /* Generate primitive Verilog modules, which are corner stones of FPGA fabric
    * Note that this function MUST be called before Verilog generation of
    * core logic (i.e., logic blocks and routing resources) !!!
@@ -121,4 +124,106 @@ void fpga_fabric_verilog(ModuleManager& module_manager,
            module_manager.num_modules());  
 }
 
+
+/********************************************************************
+ * A top-level function of FPGA-Verilog which focuses on fabric Verilog generation
+ * This function will generate
+ *  - A wrapper module, which encapsulate the FPGA module in a Verilog module which have the same port as the input benchmark 
+ *  - Testbench, where a FPGA module is configured with a bitstream and then driven by input vectors
+ *  - Pre-configured testbench, which can skip the configuration phase and pre-configure the FPGA module.
+ *    This testbench is created for quick verification and formal verification purpose.
+ *  - Verilog netlist including preprocessing flags and all the Verilog netlists that have been generated
+ ********************************************************************/
+void fpga_verilog_testbench(const ModuleManager& module_manager,
+                            const BitstreamManager& bitstream_manager, 
+                            const std::vector<ConfigBitId>& fabric_bitstream, 
+                            const AtomContext& atom_ctx, 
+                            const PlacementContext& place_ctx, 
+                            const IoLocationMap& io_location_map,
+                            const VprNetlistAnnotation& netlist_annotation, 
+                            const CircuitLibrary& circuit_lib,
+                            const SimulationSetting& simulation_setting,
+                            const e_config_protocol_type& config_protocol_type,
+                            const VerilogTestbenchOption& options) {
+
+  vtr::ScopedStartFinishTimer timer("Write Verilog testbenches for FPGA fabric\n");
+
+  std::string src_dir_path = format_dir_path(options.output_directory());
+
+  std::string netlist_name = atom_ctx.nlist.netlist_name();
+
+  /* Create directories */
+  create_dir_path(src_dir_path.c_str());
+
+  /* TODO: check if this works here. This function was in fabric generator */
+  print_verilog_simulation_preprocessing_flags(std::string(src_dir_path),
+                                               options);
+
+  /* Collect global ports from the circuit library:
+   * TODO: should we place this in the OpenFPGA context?
+   */
+  std::vector<CircuitPortId> global_ports = find_circuit_library_global_ports(circuit_lib);
+
+  /* Generate wrapper module for FPGA fabric (mapped by the input benchmark and pre-configured testbench for verification */
+  if (true == options.print_formal_verification_top_netlist()) {
+    std::string formal_verification_top_netlist_file_path = src_dir_path + netlist_name 
+                                                          + std::string(FORMAL_VERIFICATION_VERILOG_FILE_POSTFIX);
+    print_verilog_preconfig_top_module(module_manager, bitstream_manager,
+                                       circuit_lib, global_ports,
+                                       atom_ctx, place_ctx, io_location_map,
+                                       netlist_annotation,
+                                       netlist_name,
+                                       formal_verification_top_netlist_file_path,
+                                       src_dir_path);
+  }
+
+  if (true == options.print_preconfig_top_testbench()) {
+    /* Generate top-level testbench using random vectors */
+    std::string random_top_testbench_file_path = src_dir_path + netlist_name 
+                                               + std::string(RANDOM_TOP_TESTBENCH_VERILOG_FILE_POSTFIX);
+    print_verilog_random_top_testbench(netlist_name,
+                                       random_top_testbench_file_path, 
+                                       src_dir_path,
+                                       atom_ctx,  
+                                       netlist_annotation,
+                                       simulation_setting);
+  }
+
+  /* Generate full testbench for verification, including configuration phase and operating phase */
+  if (true == options.print_top_testbench()) {
+    std::string top_testbench_file_path = src_dir_path + netlist_name
+                                        + std::string(AUTOCHECK_TOP_TESTBENCH_VERILOG_FILE_POSTFIX);
+    print_verilog_top_testbench(module_manager,
+                                bitstream_manager, fabric_bitstream,
+                                config_protocol_type,
+                                circuit_lib, global_ports,
+                                atom_ctx, place_ctx, io_location_map,
+                                netlist_annotation,
+                                netlist_name,
+                                top_testbench_file_path,
+                                src_dir_path,
+                                simulation_setting);
+  }
+ 
+  /* Generate exchangeable files which contains simulation settings */
+  if (true == options.print_simulation_ini()) {
+    std::string simulation_ini_file_name = options.simulation_ini_path();
+    VTR_ASSERT(true != options.simulation_ini_path().empty());
+    print_verilog_simulation_info(simulation_ini_file_name,
+                                  netlist_name,
+                                  src_dir_path,
+                                  bitstream_manager.bits().size(),
+                                  simulation_setting.num_clock_cycles(),
+                                  simulation_setting.programming_clock_frequency(),
+                                  simulation_setting.operating_clock_frequency());
+  }
+
+  /* Generate a Verilog file including all the netlists that have been generated */
+  print_include_netlists(src_dir_path,
+                         netlist_name,
+                         options.reference_benchmark_file_path(),
+                         circuit_lib);
+
+}
+
 } /* end namespace openfpga */

openfpga/src/fpga_verilog/verilog_api.h

@@ -7,14 +7,18 @@
 
 #include <string>
 #include <vector>
-#include "vpr_types.h"
 #include "mux_library.h"
 #include "circuit_library.h"
 #include "vpr_context.h"
 #include "vpr_device_annotation.h"
 #include "device_rr_gsb.h"
 #include "module_manager.h"
-#include "verilog_options.h"
+#include "bitstream_manager.h"
+#include "simulation_setting.h"
+#include "io_location_map.h"
+#include "vpr_netlist_annotation.h"
+#include "fabric_verilog_options.h"
+#include "verilog_testbench_options.h"
 
 /********************************************************************
  * Function declaration
@@ -31,6 +35,19 @@ void fpga_fabric_verilog(ModuleManager& module_manager,
                          const DeviceRRGSB& device_rr_gsb,
                          const FabricVerilogOption& options);
 
+void fpga_verilog_testbench(const ModuleManager& module_manager,
+                            const BitstreamManager& bitstream_manager, 
+                            const std::vector<ConfigBitId>& fabric_bitstream, 
+                            const AtomContext& atom_ctx, 
+                            const PlacementContext& place_ctx, 
+                            const IoLocationMap& io_location_map,
+                            const VprNetlistAnnotation& netlist_annotation, 
+                            const CircuitLibrary& circuit_lib,
+                            const SimulationSetting& simulation_parameters,
+                            const e_config_protocol_type& config_protocol_type,
+                            const VerilogTestbenchOption& options);
+
+
 } /* end namespace openfpga */
 
 #endif

openfpga/src/fpga_verilog/verilog_auxiliary_netlists.cpp

@@ -23,7 +23,6 @@ namespace openfpga {
 /********************************************************************
  * Local constant variables
  *******************************************************************/
-constexpr char* TOP_INCLUDE_NETLIST_FILE_NAME_POSTFIX = "_include_netlists.v";
 
 /********************************************************************
  * Print a file that includes all the netlists that have been generated
@@ -115,7 +114,7 @@ void print_include_netlists(const std::string& src_dir,
  * which are used enable/disable some features in FPGA Verilog modules
  *******************************************************************/
 void print_verilog_preprocessing_flags_netlist(const std::string& src_dir,
-                                               const FabricVerilogOption& fpga_verilog_opts) {
+                                               const FabricVerilogOption& fabric_verilog_opts) {
 
   std::string verilog_fname = src_dir + std::string(DEFINES_VERILOG_FILE_NAME);
 
@@ -130,25 +129,19 @@ void print_verilog_preprocessing_flags_netlist(const std::string& src_dir,
   print_verilog_file_header(fp, std::string("Preprocessing flags to enable/disable features in FPGA Verilog modules")); 
 
   /* To enable timing */
-  if (true == fpga_verilog_opts.include_timing()) {
+  if (true == fabric_verilog_opts.include_timing()) {
     print_verilog_define_flag(fp, std::string(VERILOG_TIMING_PREPROC_FLAG), 1);
     fp << std::endl;
   } 
 
   /* To enable timing */
-  if (true == fpga_verilog_opts.include_signal_init()) {
+  if (true == fabric_verilog_opts.include_signal_init()) {
     print_verilog_define_flag(fp, std::string(VERILOG_SIGNAL_INIT_PREPROC_FLAG), 1);
     fp << std::endl;
   } 
 
-  /* To enable formal verfication flag */
-  if (true == fpga_verilog_opts.print_formal_verification_top_netlist()) {
-    print_verilog_define_flag(fp, std::string(VERILOG_FORMAL_VERIFICATION_PREPROC_FLAG), 1);
-    fp << std::endl;
-  } 
-
   /* To enable functional verfication with Icarus */
-  if (true == fpga_verilog_opts.support_icarus_simulator()) {
+  if (true == fabric_verilog_opts.support_icarus_simulator()) {
     print_verilog_define_flag(fp, std::string(ICARUS_SIMULATOR_FLAG), 1);
     fp << std::endl;
   } 
@@ -161,7 +154,7 @@ void print_verilog_preprocessing_flags_netlist(const std::string& src_dir,
  * Print a Verilog file containing simulation-related preprocessing flags
  *******************************************************************/
 void print_verilog_simulation_preprocessing_flags(const std::string& src_dir,
-                                                  const FabricVerilogOption& fpga_verilog_opts) {
+                                                  const VerilogTestbenchOption& verilog_testbench_opts) {
 
   std::string verilog_fname = src_dir + std::string(DEFINES_VERILOG_SIMULATION_FILE_NAME);
 
@@ -176,19 +169,26 @@ void print_verilog_simulation_preprocessing_flags(const std::string& src_dir,
   print_verilog_file_header(fp, std::string("Preprocessing flags to enable/disable simulation features")); 
 
   /* To enable manualy checked simulation */
-  if (true == fpga_verilog_opts.print_top_testbench()) {
+  if (true == verilog_testbench_opts.print_top_testbench()) {
     print_verilog_define_flag(fp, std::string(INITIAL_SIMULATION_FLAG), 1);
     fp << std::endl;
   } 
 
   /* To enable auto-checked simulation */
-  if (true == fpga_verilog_opts.print_autocheck_top_testbench()) {
+  if ( (true == verilog_testbench_opts.print_preconfig_top_testbench())
+    || (true == verilog_testbench_opts.print_top_testbench()) ) {
     print_verilog_define_flag(fp, std::string(AUTOCHECKED_SIMULATION_FLAG), 1);
     fp << std::endl;
   } 
 
   /* To enable pre-configured FPGA simulation */
-  if (true == fpga_verilog_opts.print_formal_verification_top_netlist()) {
+  if (true == verilog_testbench_opts.print_formal_verification_top_netlist()) {
+    print_verilog_define_flag(fp, std::string(VERILOG_FORMAL_VERIFICATION_PREPROC_FLAG), 1);
+    fp << std::endl;
+  } 
+
+  /* To enable pre-configured FPGA simulation */
+  if (true == verilog_testbench_opts.print_preconfig_top_testbench()) {
     print_verilog_define_flag(fp, std::string(FORMAL_SIMULATION_FLAG), 1);
     fp << std::endl;
   } 

openfpga/src/fpga_verilog/verilog_auxiliary_netlists.h

@@ -6,7 +6,8 @@
  *******************************************************************/
 #include <string>
 #include "circuit_library.h"
-#include "verilog_options.h"
+#include "fabric_verilog_options.h"
+#include "verilog_testbench_options.h"
 
 /********************************************************************
  * Function declaration
@@ -21,10 +22,10 @@ void print_include_netlists(const std::string& src_dir,
                             const CircuitLibrary& circuit_lib);
 
 void print_verilog_preprocessing_flags_netlist(const std::string& src_dir,
-                                               const FabricVerilogOption& fpga_verilog_opts);
+                                               const FabricVerilogOption& fabric_verilog_opts);
 
 void print_verilog_simulation_preprocessing_flags(const std::string& src_dir,
-                                                  const FabricVerilogOption& fpga_verilog_opts);
+                                                  const VerilogTestbenchOption& verilog_testbench_opts);
 
 } /* end namespace openfpga */
 

openfpga/src/fpga_verilog/verilog_constants.h

@@ -22,6 +22,7 @@ constexpr char* MODELSIM_SIMULATION_TIME_UNIT = "ms";
 constexpr char* ICARUS_SIMULATOR_FLAG = "ICARUS_SIMULATOR"; // the flag to enable specific Verilog code in testbenches
 // End of Icarus variables and flag
 
+constexpr char* TOP_INCLUDE_NETLIST_FILE_NAME_POSTFIX = "_include_netlists.v";
 constexpr char* VERILOG_TOP_POSTFIX = "_top.v";
 constexpr char* FORMAL_VERIFICATION_VERILOG_FILE_POSTFIX = "_top_formal_verification.v"; 
 constexpr char* TOP_TESTBENCH_VERILOG_FILE_POSTFIX = "_top_tb.v"; /* !!! must be consist with the modelsim_testbench_module_postfix */ 
@@ -48,4 +49,12 @@ constexpr char* SB_VERILOG_FILE_NAME_PREFIX = "sb_";
 constexpr char* LOGICAL_MODULE_VERILOG_FILE_NAME_PREFIX = "logical_tile_";
 constexpr char* GRID_VERILOG_FILE_NAME_PREFIX = "grid_";
 
+constexpr char* FORMAL_VERIFICATION_TOP_MODULE_POSTFIX = "_top_formal_verification";
+constexpr char* FORMAL_VERIFICATION_TOP_MODULE_PORT_POSTFIX = "_fm";
+constexpr char* FORMAL_VERIFICATION_TOP_MODULE_UUT_NAME = "U0_formal_verification";
+
+constexpr char* FORMAL_RANDOM_TOP_TESTBENCH_POSTFIX = "_top_formal_verification_random_tb";
+
+#define VERILOG_DEFAULT_SIGNAL_INIT_VALUE 0
+
 #endif

openfpga/src/fpga_verilog/verilog_formal_random_top_testbench.cpp

@@ -0,0 +1,268 @@
+/********************************************************************
+ * This file includes functions that are used to generate a Verilog 
+ * testbench for the top-level module (FPGA fabric), in purpose of
+ * running formal verification with random input vectors 
+ *******************************************************************/
+#include <fstream>
+#include <cstring>
+#include <cmath>
+#include <iomanip>
+
+/* Headers from vtrutil library */
+#include "vtr_assert.h"
+#include "vtr_log.h"
+#include "vtr_time.h"
+
+/* Headers from openfpgautil library */
+#include "openfpga_port.h"
+#include "openfpga_digest.h"
+
+#include "openfpga_atom_netlist_utils.h"
+#include "simulation_utils.h"
+
+#include "verilog_constants.h"
+#include "verilog_writer_utils.h"
+#include "verilog_testbench_utils.h"
+#include "verilog_formal_random_top_testbench.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * Local variables used only in this file
+ *******************************************************************/
+constexpr char* FPGA_PORT_POSTFIX = "_gfpga";
+constexpr char* BENCHMARK_PORT_POSTFIX = "_bench";
+constexpr char* CHECKFLAG_PORT_POSTFIX = "_flag";
+constexpr char* DEFAULT_CLOCK_NAME = "clk";
+constexpr char* BENCHMARK_INSTANCE_NAME = "REF_DUT";
+constexpr char* FPGA_INSTANCE_NAME = "FPGA_DUT";
+constexpr char* ERROR_COUNTER = "nb_error";
+constexpr char* FORMAL_TB_SIM_START_PORT_NAME = "sim_start";
+constexpr int MAGIC_NUMBER_FOR_SIMULATION_TIME = 200;
+
+/********************************************************************
+ * Print the module ports for the Verilog testbench 
+ * using random vectors 
+ * This function generates 
+ * 1. the input ports to drive both input benchmark module and FPGA fabric module
+ * 2. the output ports for input benchmark module
+ * 3. the output ports for FPGA fabric module
+ * 4. the error checking ports 
+ *******************************************************************/
+static 
+void print_verilog_top_random_testbench_ports(std::fstream& fp,
+                                              const std::string& circuit_name,
+                                              const std::vector<std::string>& clock_port_names,
+                                              const AtomContext& atom_ctx,
+                                              const VprNetlistAnnotation& netlist_annotation) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+ 
+  /* Print the declaration for the module */
+  fp << "module " << circuit_name << FORMAL_RANDOM_TOP_TESTBENCH_POSTFIX << ";" << std::endl;
+
+  /* Create a clock port if the benchmark does not have one! 
+   * The clock is used for counting and synchronizing input stimulus 
+   */
+  BasicPort clock_port = generate_verilog_testbench_clock_port(clock_port_names, std::string(DEFAULT_CLOCK_NAME));
+  print_verilog_comment(fp, std::string("----- Default clock port is added here since benchmark does not contain one -------"));
+  fp << "\t" << generate_verilog_port(VERILOG_PORT_REG, clock_port) << ";" << std::endl;
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+
+  print_verilog_testbench_shared_ports(fp, atom_ctx, netlist_annotation,
+                                       clock_port_names,
+                                       std::string(BENCHMARK_PORT_POSTFIX),
+                                       std::string(FPGA_PORT_POSTFIX),
+                                       std::string(CHECKFLAG_PORT_POSTFIX),
+                                       std::string(AUTOCHECKED_SIMULATION_FLAG));
+
+  /* Instantiate an integer to count the number of error 
+   * and determine if the simulation succeed or failed 
+   */
+  print_verilog_comment(fp, std::string("----- Error counter -------"));
+  fp << "\tinteger " << ERROR_COUNTER << "= 0;" << std::endl;
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Instanciate the input benchmark module
+ *******************************************************************/
+static
+void print_verilog_top_random_testbench_benchmark_instance(std::fstream& fp, 
+                                                           const std::string& reference_verilog_top_name,
+                                                           const AtomContext& atom_ctx,
+                                                           const VprNetlistAnnotation& netlist_annotation) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  /* Benchmark is instanciated conditionally: only when a preprocessing flag is enable */
+  print_verilog_preprocessing_flag(fp, std::string(AUTOCHECKED_SIMULATION_FLAG)); 
+
+  print_verilog_comment(fp, std::string("----- Reference Benchmark Instanication -------"));
+
+  /* Do NOT use explicit port mapping here: 
+   * VPR added a prefix of "out_" to the output ports of input benchmark
+   */
+  print_verilog_testbench_benchmark_instance(fp, reference_verilog_top_name,
+                                             std::string(BENCHMARK_INSTANCE_NAME),
+                                             std::string(),
+                                             std::string(),
+                                             std::string(BENCHMARK_PORT_POSTFIX),
+                                             atom_ctx, netlist_annotation,
+                                             false);
+
+  print_verilog_comment(fp, std::string("----- End reference Benchmark Instanication -------"));
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+
+  /* Condition ends for the benchmark instanciation */
+  print_verilog_endif(fp);
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Instanciate the FPGA fabric module
+ *******************************************************************/
+static
+void print_verilog_random_testbench_fpga_instance(std::fstream& fp,
+                                                  const std::string& circuit_name,
+                                                  const AtomContext& atom_ctx,
+                                                  const VprNetlistAnnotation& netlist_annotation) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  print_verilog_comment(fp, std::string("----- FPGA fabric instanciation -------"));
+
+  /* Always use explicit port mapping */
+  print_verilog_testbench_benchmark_instance(fp, std::string(circuit_name + std::string(FORMAL_VERIFICATION_TOP_MODULE_POSTFIX)),
+                                             std::string(FPGA_INSTANCE_NAME),
+                                             std::string(FORMAL_VERIFICATION_TOP_MODULE_PORT_POSTFIX),
+                                             std::string(FORMAL_VERIFICATION_TOP_MODULE_PORT_POSTFIX),
+                                             std::string(FPGA_PORT_POSTFIX),
+                                             atom_ctx, netlist_annotation,
+                                             true);
+
+  print_verilog_comment(fp, std::string("----- End FPGA Fabric Instanication -------"));
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+}
+
+/*********************************************************************
+ * Top-level function in this file:
+ * Create a Verilog testbench using random input vectors 
+ * The testbench consists of two modules, i.e., Design Under Test (DUT)
+ * 1. top-level module of FPGA fabric
+ * 2. top-level module of users' benchmark, 
+ *    i.e., the input benchmark of VPR flow
+ *                                    +----------+
+ *                                    |   FPGA   |       +------------+
+ *                             +----->|  Fabric  |------>|            |
+ *                             |      |          |       |            |
+ *                             |      +----------+       |            |
+ *                             |                         | Output     | 
+ *   random_input_vectors -----+                         | Vector     |---->Functional correct?
+ *                             |                         | Comparator |    
+ *                             |      +-----------+      |            |
+ *                             |      |  Input    |      |            |
+ *                             +----->| Benchmark |----->|            |
+ *                                    +-----------+      +------------+
+ *
+ * Same input vectors are given to drive both DUTs.
+ * The output vectors of the DUTs are compared to verify if they
+ * have the same functionality.
+ * A flag will be raised to indicate the result 
+ ********************************************************************/
+void print_verilog_random_top_testbench(const std::string& circuit_name,
+                                        const std::string& verilog_fname,
+                                        const std::string& verilog_dir,
+                                        const AtomContext& atom_ctx,
+                                        const VprNetlistAnnotation& netlist_annotation,
+                                        const SimulationSetting& simulation_parameters) {
+  std::string timer_message = std::string("Write configuration-skip testbench for FPGA top-level Verilog netlist implemented by '") + circuit_name.c_str() + std::string("'");
+
+  /* Start time count */
+  vtr::ScopedStartFinishTimer timer(timer_message);
+
+  /* Create the file stream */
+  std::fstream fp;
+  fp.open(verilog_fname, std::fstream::out | std::fstream::trunc);
+
+  /* Validate the file stream */
+  check_file_stream(verilog_fname.c_str(), fp);
+
+  /* Generate a brief description on the Verilog file*/
+  std::string title = std::string("FPGA Verilog Testbench for Formal Top-level netlist of Design: ") + circuit_name;
+  print_verilog_file_header(fp, title); 
+
+  /* Print preprocessing flags and external netlists */
+  print_verilog_include_defines_preproc_file(fp, verilog_dir);
+
+  print_verilog_include_netlist(fp, std::string(verilog_dir + std::string(DEFINES_VERILOG_SIMULATION_FILE_NAME)));  
+
+  /* Preparation: find all the clock ports */
+  std::vector<std::string> clock_port_names = find_atom_netlist_clock_port_names(atom_ctx.nlist, netlist_annotation);
+
+  /* Start of testbench */
+  print_verilog_top_random_testbench_ports(fp, circuit_name, clock_port_names, atom_ctx, netlist_annotation);
+
+  /* Call defined top-level module */
+  print_verilog_random_testbench_fpga_instance(fp, circuit_name, atom_ctx, netlist_annotation);
+
+  /* Call defined benchmark */
+  print_verilog_top_random_testbench_benchmark_instance(fp, circuit_name, atom_ctx, netlist_annotation);
+
+  /* Find clock port to be used */
+  BasicPort clock_port = generate_verilog_testbench_clock_port(clock_port_names, std::string(DEFAULT_CLOCK_NAME));
+
+  /* Add stimuli for reset, set, clock and iopad signals */
+  print_verilog_testbench_clock_stimuli(fp, simulation_parameters, 
+                                        clock_port);
+  print_verilog_testbench_random_stimuli(fp, atom_ctx,
+                                         netlist_annotation, 
+                                         clock_port_names, 
+                                         std::string(CHECKFLAG_PORT_POSTFIX),
+                                         clock_port);
+
+  print_verilog_testbench_check(fp, 
+                                std::string(AUTOCHECKED_SIMULATION_FLAG),
+                                std::string(FORMAL_TB_SIM_START_PORT_NAME),
+                                std::string(BENCHMARK_PORT_POSTFIX),
+                                std::string(FPGA_PORT_POSTFIX),
+                                std::string(CHECKFLAG_PORT_POSTFIX),
+                                std::string(ERROR_COUNTER),
+                                atom_ctx,
+                                netlist_annotation, 
+                                clock_port_names,
+                                std::string(DEFAULT_CLOCK_NAME));
+
+  int simulation_time = find_operating_phase_simulation_time(MAGIC_NUMBER_FOR_SIMULATION_TIME,
+                                                             simulation_parameters.num_clock_cycles(),
+                                                             1./simulation_parameters.operating_clock_frequency(),
+                                                             VERILOG_SIM_TIMESCALE);
+
+  /* Add Icarus requirement */
+  print_verilog_timeout_and_vcd(fp, 
+                                std::string(ICARUS_SIMULATOR_FLAG),
+                                std::string(circuit_name + std::string(FORMAL_RANDOM_TOP_TESTBENCH_POSTFIX)),
+                                std::string(circuit_name + std::string("_formal.vcd")), 
+                                std::string(FORMAL_TB_SIM_START_PORT_NAME),
+                                std::string(ERROR_COUNTER),
+                                simulation_time);
+
+  /* Testbench ends*/
+  print_verilog_module_end(fp, std::string(circuit_name) + std::string(FORMAL_RANDOM_TOP_TESTBENCH_POSTFIX));
+
+  /* Close the file stream */
+  fp.close();
+}
+
+} /* end namespace openfpga */

openfpga/src/fpga_verilog/verilog_formal_random_top_testbench.h

@@ -0,0 +1,27 @@
+#ifndef VERILOG_FORMAL_RANDOM_TOP_TESTBENCH
+#define VERILOG_FORMAL_RANDOM_TOP_TESTBENCH
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+#include <string>
+#include "vpr_context.h"
+#include "simulation_setting.h"
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+void print_verilog_random_top_testbench(const std::string& circuit_name,
+                                        const std::string& verilog_fname,
+                                        const std::string& verilog_dir,
+                                        const AtomContext& atom_ctx,
+                                        const VprNetlistAnnotation& netlist_annotation,
+                                        const SimulationSetting& simulation_parameters);
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/fpga_verilog/verilog_preconfig_top_module.cpp

@@ -0,0 +1,452 @@
+/********************************************************************
+ * This file includes functions that are used to generate 
+ * a Verilog module of a pre-configured FPGA fabric 
+ *******************************************************************/
+#include <fstream>
+
+/* Headers from vtrutil library */
+#include "vtr_assert.h"
+#include "vtr_log.h"
+#include "vtr_time.h"
+
+/* Headers from openfpgautil library */
+#include "openfpga_port.h"
+#include "openfpga_digest.h"
+
+#include "bitstream_manager_utils.h"
+#include "openfpga_atom_netlist_utils.h"
+
+#include "openfpga_naming.h"
+
+#include "verilog_constants.h"
+#include "verilog_writer_utils.h"
+#include "verilog_testbench_utils.h"
+#include "verilog_preconfig_top_module.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * Print module declaration and ports for the pre-configured
+ * FPGA top module
+ * The module ports do exactly match the input benchmark 
+ *******************************************************************/
+static 
+void print_verilog_preconfig_top_module_ports(std::fstream& fp, 
+                                              const std::string& circuit_name,
+                                              const AtomContext& atom_ctx,
+                                              const VprNetlistAnnotation& netlist_annotation) {
+
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  /* Module declaration */ 
+  fp << "module " << circuit_name << std::string(FORMAL_VERIFICATION_TOP_MODULE_POSTFIX);
+  fp << " (" << std::endl;
+ 
+  /* Add module ports */
+  size_t port_counter = 0;
+
+  /* Port type-to-type mapping */
+  std::map<AtomBlockType, enum e_dump_verilog_port_type> port_type2type_map;
+  port_type2type_map[AtomBlockType::INPAD] = VERILOG_PORT_INPUT;
+  port_type2type_map[AtomBlockType::OUTPAD] = VERILOG_PORT_OUTPUT;
+  
+  /* Print all the I/Os of the circuit implementation to be tested*/
+  for (const AtomBlockId& atom_blk : atom_ctx.nlist.blocks()) {
+    /* We only care I/O logical blocks !*/
+    if ( (AtomBlockType::INPAD != atom_ctx.nlist.block_type(atom_blk))
+      && (AtomBlockType::OUTPAD != atom_ctx.nlist.block_type(atom_blk)) ) {
+      continue;
+    }
+
+    /* The block may be renamed as it contains special characters which violate Verilog syntax */
+    std::string block_name = atom_ctx.nlist.block_name(atom_blk);
+    if (true == netlist_annotation.is_block_renamed(atom_blk)) {
+      block_name = netlist_annotation.block_name(atom_blk);
+    } 
+
+    if (0 < port_counter) { 
+      fp << "," << std::endl;
+    }
+    /* Both input and output ports have only size of 1 */
+    BasicPort module_port(std::string(block_name + std::string(FORMAL_VERIFICATION_TOP_MODULE_PORT_POSTFIX)), 1); 
+    fp << generate_verilog_port(port_type2type_map[atom_ctx.nlist.block_type(atom_blk)], module_port);  
+
+    /* Update port counter */
+    port_counter++;
+  }
+
+  fp << ");" << std::endl;
+
+  /* Add an empty line as a splitter */
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Print internal wires for the pre-configured FPGA top module
+ * The internal wires are tailored for the ports of FPGA top module
+ * which will be different in various configuration protocols
+ *******************************************************************/
+static 
+void print_verilog_preconfig_top_module_internal_wires(std::fstream& fp,
+                                                       const ModuleManager& module_manager,
+                                                       const ModuleId& top_module) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  /* Global ports of top-level module  */
+  print_verilog_comment(fp, std::string("----- Local wires for FPGA fabric -----"));
+  for (const ModulePortId& module_port_id : module_manager.module_ports(top_module)) {
+    BasicPort module_port = module_manager.module_port(top_module, module_port_id);
+    fp << generate_verilog_port(VERILOG_PORT_WIRE, module_port) << ";" << std::endl;
+  }
+  /* Add an empty line as a splitter */
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Connect global ports of FPGA top module to constants except:
+ * 1. operating clock, which should be wired to the clock port of 
+ * this pre-configured FPGA top module
+ *******************************************************************/
+static 
+void print_verilog_preconfig_top_module_connect_global_ports(std::fstream& fp,
+                                                             const ModuleManager& module_manager,
+                                                             const ModuleId& top_module,
+                                                             const CircuitLibrary& circuit_lib,
+                                                             const std::vector<CircuitPortId>& global_ports,
+                                                             const std::vector<std::string>& benchmark_clock_port_names) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  print_verilog_comment(fp, std::string("----- Begin Connect Global ports of FPGA top module -----"));
+
+  /* Global ports of the top module in module manager do not carry any attributes, 
+   * such as is_clock, is_set, etc. 
+   * Therefore, for each global port in the top module, we find the circuit port in the circuit library
+   * which share the same name. We can access to the attributes.
+   * To gurantee the correct link between global ports in module manager and those in circuit library
+   * We have performed some critical check in check_circuit_library() for global ports,
+   * where we guarantee all the global ports share the same name must have the same attributes.
+   * So that each global port with the same name is unique!
+   */
+  for (const BasicPort& module_global_port : module_manager.module_ports_by_type(top_module, ModuleManager::MODULE_GLOBAL_PORT)) {
+    CircuitPortId linked_circuit_port_id = CircuitPortId::INVALID();
+    /* Find the circuit port with the same name */
+    for (const CircuitPortId& circuit_port_id : global_ports) {
+      if (0 != module_global_port.get_name().compare(circuit_lib.port_prefix(circuit_port_id))) {
+        continue;
+      }
+      linked_circuit_port_id = circuit_port_id;
+      break;
+    }
+    /* Must find one valid circuit port */
+    VTR_ASSERT(CircuitPortId::INVALID() != linked_circuit_port_id);
+    /* Port size should match! */
+    VTR_ASSERT(module_global_port.get_width() == circuit_lib.port_size(linked_circuit_port_id));
+    /* Now, for operating clock port, we should wire it to the clock of benchmark! */
+    if  ( (CIRCUIT_MODEL_PORT_CLOCK == circuit_lib.port_type(linked_circuit_port_id)) 
+       && (false == circuit_lib.port_is_prog(linked_circuit_port_id)) ) {
+      /* Wiring to each pin of the global port: benchmark clock is always 1-bit */
+      for (const size_t& pin : module_global_port.pins()) {  
+        for (const std::string& clock_port_name : benchmark_clock_port_names) {  
+          BasicPort module_clock_pin(module_global_port.get_name(), pin, pin);
+          BasicPort benchmark_clock_pin(clock_port_name + std::string(FORMAL_VERIFICATION_TOP_MODULE_PORT_POSTFIX), 1);
+          print_verilog_wire_connection(fp, module_clock_pin, benchmark_clock_pin, false);
+        }
+      } 
+      /* Finish, go to the next */
+      continue;
+    } 
+
+    /* For other ports, give an default value */
+    std::vector<size_t> default_values(module_global_port.get_width(), circuit_lib.port_default_value(linked_circuit_port_id)); 
+    print_verilog_wire_constant_values(fp, module_global_port, default_values);
+  }
+
+  print_verilog_comment(fp, std::string("----- End Connect Global ports of FPGA top module -----"));
+
+  /* Add an empty line as a splitter */
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Impose the bitstream on the configuration memories
+ * This function uses 'assign' syntax to impost the bitstream at mem port
+ * while uses 'force' syntax to impost the bitstream at mem_inv port
+ *******************************************************************/
+static 
+void print_verilog_preconfig_top_module_assign_bitstream(std::fstream& fp,
+                                                         const ModuleManager& module_manager,
+                                                         const ModuleId& top_module,
+                                                         const BitstreamManager& bitstream_manager) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  print_verilog_comment(fp, std::string("----- Begin assign bitstream to configuration memories -----"));
+
+  for (const ConfigBlockId& config_block_id : bitstream_manager.blocks()) {
+    /* We only cares blocks with configuration bits */
+    if (0 == bitstream_manager.block_bits(config_block_id).size()) {
+      continue;
+    }
+    /* Build the hierarchical path of the configuration bit in modules */   
+    std::vector<ConfigBlockId> block_hierarchy = find_bitstream_manager_block_hierarchy(bitstream_manager, config_block_id);
+    /* Drop the first block, which is the top module, it should be replaced by the instance name here */
+    /* Ensure that this is the module we want to drop! */
+    VTR_ASSERT(0 == module_manager.module_name(top_module).compare(bitstream_manager.block_name(block_hierarchy[0])));
+    block_hierarchy.erase(block_hierarchy.begin());
+    /* Build the full hierarchy path */
+    std::string bit_hierarchy_path(FORMAL_VERIFICATION_TOP_MODULE_UUT_NAME);
+    for (const ConfigBlockId& temp_block : block_hierarchy) {
+      bit_hierarchy_path += std::string(".");
+      bit_hierarchy_path += bitstream_manager.block_name(temp_block);
+    }
+    bit_hierarchy_path += std::string(".");
+
+    /* Find the bit index in the parent block */
+    BasicPort config_data_port(bit_hierarchy_path + generate_configuration_chain_data_out_name(),
+                               bitstream_manager.block_bits(config_block_id).size());
+
+    /* Wire it to the configuration bit: access both data out and data outb ports */
+    std::vector<size_t> config_data_values;
+    for (const ConfigBitId config_bit : bitstream_manager.block_bits(config_block_id)) {
+      config_data_values.push_back(bitstream_manager.bit_value(config_bit));
+    }
+    print_verilog_wire_constant_values(fp, config_data_port, config_data_values);
+  }
+
+  fp << "initial begin" << std::endl;
+
+  for (const ConfigBlockId& config_block_id : bitstream_manager.blocks()) {
+    /* We only cares blocks with configuration bits */
+    if (0 == bitstream_manager.block_bits(config_block_id).size()) {
+      continue;
+    }
+    /* Build the hierarchical path of the configuration bit in modules */   
+    std::vector<ConfigBlockId> block_hierarchy = find_bitstream_manager_block_hierarchy(bitstream_manager, config_block_id);
+    /* Drop the first block, which is the top module, it should be replaced by the instance name here */
+    /* Ensure that this is the module we want to drop! */
+    VTR_ASSERT(0 == module_manager.module_name(top_module).compare(bitstream_manager.block_name(block_hierarchy[0])));
+    block_hierarchy.erase(block_hierarchy.begin());
+    /* Build the full hierarchy path */
+    std::string bit_hierarchy_path(FORMAL_VERIFICATION_TOP_MODULE_UUT_NAME);
+    for (const ConfigBlockId& temp_block : block_hierarchy) {
+      bit_hierarchy_path += std::string(".");
+      bit_hierarchy_path += bitstream_manager.block_name(temp_block);
+    }
+    bit_hierarchy_path += std::string(".");
+
+    /* Find the bit index in the parent block */
+    BasicPort config_datab_port(bit_hierarchy_path + generate_configuration_chain_inverted_data_out_name(),
+                                bitstream_manager.block_bits(config_block_id).size());
+
+    std::vector<size_t> config_datab_values;
+    for (const ConfigBitId config_bit : bitstream_manager.block_bits(config_block_id)) {
+      config_datab_values.push_back(!bitstream_manager.bit_value(config_bit));
+    }
+    print_verilog_force_wire_constant_values(fp, config_datab_port, config_datab_values);
+  }
+
+  fp << "end" << std::endl;
+
+  print_verilog_comment(fp, std::string("----- End assign bitstream to configuration memories -----"));
+}
+
+/********************************************************************
+ * Impose the bitstream on the configuration memories
+ * This function uses '$deposit' syntax to do so
+ *******************************************************************/
+static 
+void print_verilog_preconfig_top_module_deposit_bitstream(std::fstream& fp,
+                                                          const ModuleManager& module_manager,
+                                                          const ModuleId& top_module,
+                                                          const BitstreamManager& bitstream_manager) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  print_verilog_comment(fp, std::string("----- Begin deposit bitstream to configuration memories -----"));
+  
+  fp << "initial begin" << std::endl;
+
+  for (const ConfigBlockId& config_block_id : bitstream_manager.blocks()) {
+    /* We only cares blocks with configuration bits */
+    if (0 == bitstream_manager.block_bits(config_block_id).size()) {
+      continue;
+    }
+    /* Build the hierarchical path of the configuration bit in modules */   
+    std::vector<ConfigBlockId> block_hierarchy = find_bitstream_manager_block_hierarchy(bitstream_manager, config_block_id);
+    /* Drop the first block, which is the top module, it should be replaced by the instance name here */
+    /* Ensure that this is the module we want to drop! */
+    VTR_ASSERT(0 == module_manager.module_name(top_module).compare(bitstream_manager.block_name(block_hierarchy[0])));
+    block_hierarchy.erase(block_hierarchy.begin());
+    /* Build the full hierarchy path */
+    std::string bit_hierarchy_path(FORMAL_VERIFICATION_TOP_MODULE_UUT_NAME);
+    for (const ConfigBlockId& temp_block : block_hierarchy) {
+      bit_hierarchy_path += std::string(".");
+      bit_hierarchy_path += bitstream_manager.block_name(temp_block);
+    }
+    bit_hierarchy_path += std::string(".");
+
+    /* Find the bit index in the parent block */
+    BasicPort config_data_port(bit_hierarchy_path + generate_configuration_chain_data_out_name(),
+                               bitstream_manager.block_bits(config_block_id).size());
+
+    BasicPort config_datab_port(bit_hierarchy_path + generate_configuration_chain_inverted_data_out_name(),
+                               bitstream_manager.block_bits(config_block_id).size());
+
+    /* Wire it to the configuration bit: access both data out and data outb ports */
+    std::vector<size_t> config_data_values;
+    for (const ConfigBitId config_bit : bitstream_manager.block_bits(config_block_id)) {
+      config_data_values.push_back(bitstream_manager.bit_value(config_bit));
+    }
+    print_verilog_deposit_wire_constant_values(fp, config_data_port, config_data_values);
+
+    std::vector<size_t> config_datab_values;
+    for (const ConfigBitId config_bit : bitstream_manager.block_bits(config_block_id)) {
+      config_datab_values.push_back(!bitstream_manager.bit_value(config_bit));
+    }
+    print_verilog_deposit_wire_constant_values(fp, config_datab_port, config_datab_values);
+  }
+
+  fp << "end" << std::endl;
+
+  print_verilog_comment(fp, std::string("----- End deposit bitstream to configuration memories -----"));
+}
+
+/********************************************************************
+ * Impose the bitstream on the configuration memories
+ * We branch here for different simulators:
+ * 1. iVerilog Icarus prefers using 'assign' syntax to force the values
+ * 2. Mentor Modelsim prefers using '$deposit' syntax to do so
+ *******************************************************************/
+static 
+void print_verilog_preconfig_top_module_load_bitstream(std::fstream& fp,
+                                                       const ModuleManager& module_manager,
+                                                       const ModuleId& top_module,
+                                                       const BitstreamManager& bitstream_manager) {
+  print_verilog_comment(fp, std::string("----- Begin load bitstream to configuration memories -----"));
+  
+  print_verilog_preprocessing_flag(fp, std::string(ICARUS_SIMULATOR_FLAG)); 
+
+  /* Use assign syntax for Icarus simulator */
+  print_verilog_preconfig_top_module_assign_bitstream(fp, module_manager, top_module, bitstream_manager);
+
+  fp << "`else" << std::endl;
+
+  /* Use assign syntax for Icarus simulator */
+  print_verilog_preconfig_top_module_deposit_bitstream(fp, module_manager, top_module, bitstream_manager);
+
+  print_verilog_endif(fp);
+
+  print_verilog_comment(fp, std::string("----- End load bitstream to configuration memories -----"));
+}
+
+
+/********************************************************************
+ * Top-level function to generate a Verilog module of 
+ * a pre-configured FPGA fabric.
+ *
+ *   Pre-configured FPGA fabric
+ *                        +--------------------------------------------
+ *                        |
+ *                        |          FPGA fabric
+ *                        |          +-------------------------------+
+ *                        |          |                               |
+ *                        |  0/1---->|FPGA global ports              |
+ *                        |          |                               |
+ *   benchmark_clock----->|--------->|FPGA_clock                     |
+ *                        |          |                               |
+ *   benchmark_inputs---->|--------->|FPGA mapped I/Os               |
+ *                        |          |                               |
+ *   benchmark_outputs<---|<---------|FPGA mapped I/Os               |
+ *                        |          |                               |
+ *                        |  0/1---->|FPGA unmapped I/Os             |
+ *                        |          |                               |
+ *   fabric_bitstream---->|--------->|Internal_configuration_ports   |
+ *                        |          +-------------------------------+
+ *                        |
+ *                        +-------------------------------------------
+ *
+ * Note: we do NOT put this module in the module manager.
+ * Because, it is not a standard module, where we force configuration signals 
+ * This module is a wrapper for the FPGA fabric to be compatible in 
+ * the port map of input benchmark.
+ * It includes wires to force constant values to part of FPGA datapath I/Os
+ * All these are hard to implement as a module in module manager
+ *******************************************************************/
+void print_verilog_preconfig_top_module(const ModuleManager& module_manager,
+                                        const BitstreamManager& bitstream_manager,
+                                        const CircuitLibrary& circuit_lib,
+                                        const std::vector<CircuitPortId>& global_ports,
+                                        const AtomContext& atom_ctx,
+                                        const PlacementContext& place_ctx,
+                                        const IoLocationMap& io_location_map,
+                                        const VprNetlistAnnotation& netlist_annotation, 
+                                        const std::string& circuit_name,
+                                        const std::string& verilog_fname,
+                                        const std::string& verilog_dir) {
+  std::string timer_message = std::string("Write pre-configured FPGA top-level Verilog netlist for design '") + circuit_name + std::string("'");
+
+  /* Start time count */
+  vtr::ScopedStartFinishTimer timer(timer_message);
+
+  /* Create the file stream */
+  std::fstream fp;
+  fp.open(verilog_fname, std::fstream::out | std::fstream::trunc);
+
+  /* Validate the file stream */
+  check_file_stream(verilog_fname.c_str(), fp);
+
+  /* Generate a brief description on the Verilog file*/
+  std::string title = std::string("Verilog netlist for pre-configured FPGA fabric by design: ") + circuit_name;
+  print_verilog_file_header(fp, title); 
+
+  /* Print preprocessing flags and external netlists */
+  print_verilog_include_defines_preproc_file(fp, verilog_dir);
+
+  print_verilog_include_netlist(fp, std::string(verilog_dir + std::string(DEFINES_VERILOG_SIMULATION_FILE_NAME)));  
+
+  /* Print module declaration and ports */
+  print_verilog_preconfig_top_module_ports(fp, circuit_name, atom_ctx, netlist_annotation);
+
+  /* Find the top_module */
+  ModuleId top_module = module_manager.find_module(generate_fpga_top_module_name());
+  VTR_ASSERT(true == module_manager.valid_module_id(top_module));
+
+  /* Print internal wires */
+  print_verilog_preconfig_top_module_internal_wires(fp, module_manager, top_module);
+
+  /* Instanciate FPGA top-level module */
+  print_verilog_testbench_fpga_instance(fp, module_manager, top_module, 
+                                        std::string(FORMAL_VERIFICATION_TOP_MODULE_UUT_NAME)); 
+
+  /* Find clock ports in benchmark */
+  std::vector<std::string> benchmark_clock_port_names = find_atom_netlist_clock_port_names(atom_ctx.nlist, netlist_annotation);
+
+  /* Connect FPGA top module global ports to constant or benchmark global signals! */
+  print_verilog_preconfig_top_module_connect_global_ports(fp, module_manager, top_module,
+                                                          circuit_lib, global_ports,
+                                                          benchmark_clock_port_names);
+
+  /* Connect I/Os to benchmark I/Os or constant driver */
+  print_verilog_testbench_connect_fpga_ios(fp, module_manager, top_module,
+                                           atom_ctx, place_ctx, io_location_map,
+                                           netlist_annotation, 
+                                           std::string(FORMAL_VERIFICATION_TOP_MODULE_PORT_POSTFIX), 
+                                           std::string(FORMAL_VERIFICATION_TOP_MODULE_PORT_POSTFIX), 
+                                           (size_t)VERILOG_DEFAULT_SIGNAL_INIT_VALUE);
+
+  /* Assign FPGA internal SRAM/Memory ports to bitstream values */
+  print_verilog_preconfig_top_module_load_bitstream(fp, module_manager, top_module,
+                                                    bitstream_manager);
+
+  /* Testbench ends*/
+  print_verilog_module_end(fp, std::string(circuit_name) + std::string(FORMAL_VERIFICATION_TOP_MODULE_POSTFIX));
+
+  /* Close the file stream */
+  fp.close();
+}
+
+} /* end namespace openfpga */

openfpga/src/fpga_verilog/verilog_preconfig_top_module.h

@@ -0,0 +1,37 @@
+#ifndef VERILOG_PRECONFIG_TOP_MODULE_H
+#define VERILOG_PRECONFIG_TOP_MODULE_H
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+#include <vector>
+#include <string>
+#include "circuit_library.h" 
+#include "vpr_context.h" 
+#include "module_manager.h" 
+#include "bitstream_manager.h" 
+#include "io_location_map.h" 
+#include "vpr_netlist_annotation.h" 
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+void print_verilog_preconfig_top_module(const ModuleManager& module_manager,
+                                        const BitstreamManager& bitstream_manager,
+                                        const CircuitLibrary& circuit_lib,
+                                        const std::vector<CircuitPortId>& global_ports,
+                                        const AtomContext& atom_ctx,
+                                        const PlacementContext& place_ctx,
+                                        const IoLocationMap& io_location_map,
+                                        const VprNetlistAnnotation& netlist_annotation, 
+                                        const std::string& circuit_name,
+                                        const std::string& verilog_fname,
+                                        const std::string& verilog_dir);
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/fpga_verilog/verilog_submodule.h

@@ -6,7 +6,7 @@
  *******************************************************************/
 #include "module_manager.h"
 #include "mux_library.h"
-#include "verilog_options.h"
+#include "fabric_verilog_options.h"
 
 /********************************************************************
  * Function declaration

openfpga/src/fpga_verilog/verilog_testbench_options.cpp

@@ -0,0 +1,104 @@
+/******************************************************************************
+ * Memember functions for data structure VerilogTestbenchOption
+ ******************************************************************************/
+#include "vtr_assert.h"
+#include "vtr_log.h"
+
+#include "verilog_testbench_options.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/**************************************************
+ * Public Constructors
+ *************************************************/
+VerilogTestbenchOption::VerilogTestbenchOption() {
+  output_directory_.clear();
+  reference_benchmark_file_path_.clear();
+  print_preconfig_top_testbench_ = false;
+  print_formal_verification_top_netlist_ = false;
+  print_top_testbench_ = false;
+  simulation_ini_path_.clear();
+  verbose_output_ = false;
+}
+
+/**************************************************
+ * Public Accessors 
+ *************************************************/
+std::string VerilogTestbenchOption::output_directory() const {
+  return output_directory_;
+}
+
+std::string VerilogTestbenchOption::reference_benchmark_file_path() const {
+  return reference_benchmark_file_path_;
+}
+
+bool VerilogTestbenchOption::print_formal_verification_top_netlist() const {
+  return print_formal_verification_top_netlist_;
+}
+
+bool VerilogTestbenchOption::print_preconfig_top_testbench() const {
+  return print_preconfig_top_testbench_;
+}
+
+bool VerilogTestbenchOption::print_top_testbench() const {
+  return print_top_testbench_;
+}
+
+bool VerilogTestbenchOption::print_simulation_ini() const {
+  return !simulation_ini_path_.empty();
+}
+
+std::string VerilogTestbenchOption::simulation_ini_path() const {
+  return simulation_ini_path_;
+}
+
+bool VerilogTestbenchOption::verbose_output() const {
+  return verbose_output_;
+}
+
+/******************************************************************************
+ * Private Mutators
+ ******************************************************************************/
+void VerilogTestbenchOption::set_output_directory(const std::string& output_dir) {
+  output_directory_ = output_dir;
+}
+
+void VerilogTestbenchOption::set_reference_benchmark_file_path(const std::string& reference_benchmark_file_path) {
+  reference_benchmark_file_path_ = reference_benchmark_file_path;
+  /* Chain effect on other options: 
+   * Enable/disable the print_preconfig_top_testbench and print_top_testbench
+   */
+  set_print_preconfig_top_testbench(print_preconfig_top_testbench_); 
+  set_print_top_testbench(print_top_testbench_); 
+}
+ 
+void VerilogTestbenchOption::set_print_formal_verification_top_netlist(const bool& enabled) {
+  print_formal_verification_top_netlist_ = enabled;
+}
+
+void VerilogTestbenchOption::set_print_preconfig_top_testbench(const bool& enabled) {
+  print_preconfig_top_testbench_ = enabled
+                                 && (!reference_benchmark_file_path_.empty());
+  /* Enable print formal verification top_netlist if this is enabled */
+  if (true == print_preconfig_top_testbench_) {
+    if (false == print_formal_verification_top_netlist_) { 
+      VTR_LOG_WARN("Forcely enable to print top-level Verilog netlist in formal verification purpose as print pre-configured top-level Verilog testbench is enabled\n");
+      print_formal_verification_top_netlist_ = true;
+    }
+  }
+}
+
+void VerilogTestbenchOption::set_print_top_testbench(const bool& enabled) {
+  print_top_testbench_ = enabled && (!reference_benchmark_file_path_.empty());
+}
+
+void VerilogTestbenchOption::set_print_simulation_ini(const std::string& simulation_ini_path) {
+  simulation_ini_path_ = simulation_ini_path;
+}
+
+void VerilogTestbenchOption::set_verbose_output(const bool& enabled) {
+  verbose_output_ = enabled;
+}
+
+} /* end namespace openfpga */

openfpga/src/fpga_verilog/verilog_testbench_options.h

@@ -0,0 +1,62 @@
+#ifndef VERILOG_TESTBENCH_OPTIONS_H
+#define VERILOG_TESTBENCH_OPTIONS_H
+
+/********************************************************************
+ * Include header files required by the data structure definition
+ *******************************************************************/
+#include <string>
+
+/* Begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * Options for Verilog Testbench generator
+ * Typicall usage:
+ *   VerilogTestbench verilog_tb_opt();
+ *   // Set options
+ *   ...
+ *
+ *******************************************************************/
+class VerilogTestbenchOption {
+  public: /* Public constructor */
+    /* Set default options */
+    VerilogTestbenchOption();
+  public: /* Public accessors */
+    std::string output_directory() const;
+    std::string reference_benchmark_file_path() const;
+    bool print_formal_verification_top_netlist() const;
+    bool print_preconfig_top_testbench() const;
+    bool print_top_testbench() const;
+    bool print_simulation_ini() const;
+    std::string simulation_ini_path() const;
+    bool verbose_output() const;
+  public: /* Public validator */
+    bool validate() const;
+  public: /* Public mutators */
+    void set_output_directory(const std::string& output_dir);
+    /* The reference verilog file path is the key parameters that will have an impact on other options:
+     *  - print_preconfig_top_testbench
+     *  - print_top_testbench
+     * If the file path is empty, the above testbench generation will not be enabled 
+     */
+    void set_reference_benchmark_file_path(const std::string& reference_benchmark_file_path);
+    void set_print_formal_verification_top_netlist(const bool& enabled);
+    /* The preconfig top testbench generation can be enabled only when formal verification top netlist is enabled */
+    void set_print_preconfig_top_testbench(const bool& enabled);
+    void set_print_top_testbench(const bool& enabled);
+    void set_print_simulation_ini(const std::string& simulation_ini_path);
+    void set_verbose_output(const bool& enabled);
+  private: /* Internal Data */
+    std::string output_directory_;
+    std::string reference_benchmark_file_path_;
+    bool print_formal_verification_top_netlist_;
+    bool print_preconfig_top_testbench_;
+    bool print_top_testbench_;
+    /* Print simulation ini is enabled only when the path is not empty */
+    std::string simulation_ini_path_;
+    bool verbose_output_;
+};
+
+} /* End namespace openfpga*/
+
+#endif

openfpga/src/fpga_verilog/verilog_testbench_utils.cpp

@@ -0,0 +1,664 @@
+/********************************************************************
+ * This file includes most utilized functions that are used to create
+ * Verilog testbenches
+ *
+ * Note: please try to avoid using global variables in this file
+ * so that we can make it free to use anywhere
+ *******************************************************************/
+#include <algorithm>
+#include <iomanip>
+
+/* Headers from vtrutil library */
+#include "vtr_assert.h"
+
+/* Headers from openfpgautil library */
+#include "openfpga_port.h"
+#include "openfpga_digest.h"
+
+#include "verilog_port_types.h"
+
+#include "verilog_constants.h"
+#include "verilog_writer_utils.h"
+#include "verilog_testbench_utils.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * Print an instance of the FPGA top-level module
+ *******************************************************************/
+void print_verilog_testbench_fpga_instance(std::fstream& fp,
+                                           const ModuleManager& module_manager,
+                                           const ModuleId& top_module,
+                                           const std::string& top_instance_name) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  /* Include defined top-level module */
+  print_verilog_comment(fp, std::string("----- FPGA top-level module to be capsulated -----"));
+
+  /* Create an empty port-to-port name mapping, because we use default names */
+  std::map<std::string, BasicPort> port2port_name_map;
+
+  /* Use explicit port mapping for a clean instanciation */
+  print_verilog_module_instance(fp, module_manager, top_module, 
+                                top_instance_name, 
+                                port2port_name_map, true); 
+
+  /* Add an empty line as a splitter */
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Instanciate the input benchmark module
+ *******************************************************************/
+void print_verilog_testbench_benchmark_instance(std::fstream& fp,
+                                                const std::string& module_name,
+                                                const std::string& instance_name,
+                                                const std::string& module_input_port_postfix,
+                                                const std::string& module_output_port_postfix,
+                                                const std::string& output_port_postfix,
+                                                const AtomContext& atom_ctx,
+                                                const VprNetlistAnnotation& netlist_annotation,
+                                                const bool& use_explicit_port_map) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  fp << "\t" << module_name << " " << instance_name << "(" << std::endl;
+
+  size_t port_counter = 0;
+  for (const AtomBlockId& atom_blk : atom_ctx.nlist.blocks()) {
+    /* Bypass non-I/O atom blocks ! */
+    if ( (AtomBlockType::INPAD != atom_ctx.nlist.block_type(atom_blk))
+      && (AtomBlockType::OUTPAD != atom_ctx.nlist.block_type(atom_blk)) ) {
+      continue;
+    }
+
+    /* The block may be renamed as it contains special characters which violate Verilog syntax */
+    std::string block_name = atom_ctx.nlist.block_name(atom_blk);
+    if (true == netlist_annotation.is_block_renamed(atom_blk)) {
+      block_name = netlist_annotation.block_name(atom_blk);
+    } 
+
+    /* The first port does not need a comma */
+    if(0 < port_counter){
+      fp << "," << std::endl;
+    }
+    /* Input port follows the logical block name while output port requires a special postfix */
+    if (AtomBlockType::INPAD == atom_ctx.nlist.block_type(atom_blk)) {
+      fp << "\t\t";
+      if (true == use_explicit_port_map) {
+        fp << "." << block_name << module_input_port_postfix << "(";
+      }
+      fp << block_name;
+      if (true == use_explicit_port_map) {
+        fp << ")";
+      }
+    } else {
+      VTR_ASSERT_SAFE(AtomBlockType::OUTPAD == atom_ctx.nlist.block_type(atom_blk));
+      fp << "\t\t";
+      if (true == use_explicit_port_map) {
+        fp << "." << block_name << module_output_port_postfix << "(";
+      }
+      fp << block_name << output_port_postfix;
+      if (true == use_explicit_port_map) {
+        fp << ")";
+      }
+    }
+    /* Update the counter */
+    port_counter++;
+  }
+  fp << "\t);" << std::endl;
+}
+
+/********************************************************************
+ * This function adds stimuli to I/Os of FPGA fabric
+ * 1. For mapped I/Os, this function will wire them to the input ports
+ *    of the pre-configured FPGA top module
+ * 2. For unmapped I/Os, this function will assign a constant value
+ *    by default 
+ *******************************************************************/
+void print_verilog_testbench_connect_fpga_ios(std::fstream& fp,
+                                              const ModuleManager& module_manager,
+                                              const ModuleId& top_module,
+                                              const AtomContext& atom_ctx,
+                                              const PlacementContext& place_ctx,
+                                              const IoLocationMap& io_location_map,
+                                              const VprNetlistAnnotation& netlist_annotation,
+                                              const std::string& io_input_port_name_postfix,
+                                              const std::string& io_output_port_name_postfix,
+                                              const size_t& unused_io_value) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  /* In this function, we support only 1 type of I/Os */
+  VTR_ASSERT(1 == module_manager.module_ports_by_type(top_module, ModuleManager::MODULE_GPIO_PORT).size());
+  BasicPort module_io_port = module_manager.module_ports_by_type(top_module, ModuleManager::MODULE_GPIO_PORT)[0];
+
+  /* Keep tracking which I/Os have been used */
+  std::vector<bool> io_used(module_io_port.get_width(), false);
+
+  /* See if this I/O should be wired to a benchmark input/output */
+  /* Add signals from blif benchmark and short-wire them to FPGA I/O PADs
+   * This brings convenience to checking functionality  
+   */
+  print_verilog_comment(fp, std::string("----- Link BLIF Benchmark I/Os to FPGA I/Os -----"));
+
+  for (const AtomBlockId& atom_blk : atom_ctx.nlist.blocks()) {
+    /* Bypass non-I/O atom blocks ! */
+    if ( (AtomBlockType::INPAD != atom_ctx.nlist.block_type(atom_blk))
+      && (AtomBlockType::OUTPAD != atom_ctx.nlist.block_type(atom_blk)) ) {
+      continue;
+    }
+
+    /* Find the index of the mapped GPIO in top-level FPGA fabric */
+    size_t io_index = io_location_map.io_index(place_ctx.block_locs[atom_ctx.lookup.atom_clb(atom_blk)].loc.x,
+                                               place_ctx.block_locs[atom_ctx.lookup.atom_clb(atom_blk)].loc.y,
+                                               place_ctx.block_locs[atom_ctx.lookup.atom_clb(atom_blk)].loc.z);
+
+    /* Ensure that IO index is in range */
+    BasicPort module_mapped_io_port = module_io_port; 
+    /* Set the port pin index */ 
+    VTR_ASSERT(io_index < module_mapped_io_port.get_width());
+    module_mapped_io_port.set_width(io_index, io_index);
+
+    /* The block may be renamed as it contains special characters which violate Verilog syntax */
+    std::string block_name = atom_ctx.nlist.block_name(atom_blk);
+    if (true == netlist_annotation.is_block_renamed(atom_blk)) {
+      block_name = netlist_annotation.block_name(atom_blk);
+    } 
+
+    /* Create the port for benchmark I/O, due to BLIF benchmark, each I/O always has a size of 1 
+     * In addition, the input and output ports may have different postfix in naming
+     * due to verification context! Here, we give full customization on naming 
+     */
+    BasicPort benchmark_io_port;
+    if (AtomBlockType::INPAD == atom_ctx.nlist.block_type(atom_blk)) {
+      benchmark_io_port.set_name(std::string(block_name + io_input_port_name_postfix)); 
+      benchmark_io_port.set_width(1);
+      print_verilog_comment(fp, std::string("----- Blif Benchmark input " + block_name + " is mapped to FPGA IOPAD " + module_mapped_io_port.get_name() + "[" + std::to_string(io_index) + "] -----"));
+      print_verilog_wire_connection(fp, module_mapped_io_port, benchmark_io_port, false);
+    } else {
+      VTR_ASSERT(AtomBlockType::OUTPAD == atom_ctx.nlist.block_type(atom_blk));
+      benchmark_io_port.set_name(std::string(block_name + io_output_port_name_postfix)); 
+      benchmark_io_port.set_width(1);
+      print_verilog_comment(fp, std::string("----- Blif Benchmark output " + block_name + " is mapped to FPGA IOPAD " + module_mapped_io_port.get_name() + "[" + std::to_string(io_index) + "] -----"));
+      print_verilog_wire_connection(fp, benchmark_io_port, module_mapped_io_port, false);
+    }
+
+    /* Mark this I/O has been used/wired */
+    io_used[io_index] = true;
+  }
+
+  /* Add an empty line as a splitter */
+  fp << std::endl;
+
+  /* Wire the unused iopads to a constant */
+  print_verilog_comment(fp, std::string("----- Wire unused FPGA I/Os to constants -----"));
+  for (size_t io_index = 0; io_index < io_used.size(); ++io_index) {
+    /* Bypass used iopads */
+    if (true == io_used[io_index]) {
+      continue;
+    }
+
+    /* Wire to a contant */
+    BasicPort module_unused_io_port = module_io_port; 
+    /* Set the port pin index */ 
+    module_unused_io_port.set_width(io_index, io_index);
+
+    std::vector<size_t> default_values(module_unused_io_port.get_width(), unused_io_value);
+    print_verilog_wire_constant_values(fp, module_unused_io_port, default_values);
+  }
+
+  /* Add an empty line as a splitter */
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Print Verilog codes to set up a timeout for the simulation 
+ * and dump the waveform to VCD files
+ *
+ * Note that: these codes are tuned for Icarus simulator!!!
+ *******************************************************************/
+void print_verilog_timeout_and_vcd(std::fstream& fp,
+                                   const std::string& icarus_preprocessing_flag,
+                                   const std::string& module_name,
+                                   const std::string& vcd_fname,
+                                   const std::string& simulation_start_counter_name,
+                                   const std::string& error_counter_name,
+                                   const int& simulation_time) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  /* The following verilog codes are tuned for Icarus */
+  print_verilog_preprocessing_flag(fp, icarus_preprocessing_flag); 
+
+  print_verilog_comment(fp, std::string("----- Begin Icarus requirement -------"));
+
+  fp << "\tinitial begin" << std::endl;
+  fp << "\t\t$dumpfile(\"" << vcd_fname << "\");" << std::endl;
+  fp << "\t\t$dumpvars(1, " << module_name << ");" << std::endl;
+  fp << "\tend" << std::endl;
+
+  /* Condition ends for the Icarus requirement */
+  print_verilog_endif(fp);
+
+  print_verilog_comment(fp, std::string("----- END Icarus requirement -------"));
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+
+  BasicPort sim_start_port(simulation_start_counter_name, 1);
+
+  fp << "initial begin" << std::endl;
+  fp << "\t" << generate_verilog_port(VERILOG_PORT_CONKT, sim_start_port) << " <= 1'b1;" << std::endl;
+  fp << "\t$timeformat(-9, 2, \"ns\", 20);" << std::endl;
+  fp << "\t$display(\"Simulation start\");" << std::endl;
+  print_verilog_comment(fp, std::string("----- Can be changed by the user for his/her need -------"));
+  fp << "\t#" << simulation_time << std::endl;
+  fp << "\tif(" << error_counter_name << " == 0) begin" << std::endl;
+  fp << "\t\t$display(\"Simulation Succeed\");" << std::endl;
+  fp << "\tend else begin" << std::endl;
+  fp << "\t\t$display(\"Simulation Failed with " << std::string("%d") << " error(s)\", " << error_counter_name << ");" << std::endl;
+  fp << "\tend" << std::endl;
+  fp << "\t$finish;" << std::endl;
+  fp << "end" << std::endl;
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Generate the clock port name to be used in this testbench
+ *
+ * Restrictions:
+ * Assume this is a single clock benchmark
+ *******************************************************************/
+BasicPort generate_verilog_testbench_clock_port(const std::vector<std::string>& clock_port_names,
+                                                const std::string& default_clock_name) {
+  if (0 == clock_port_names.size()) {
+    return BasicPort(default_clock_name, 1); 
+  }
+
+  VTR_ASSERT(1 == clock_port_names.size());
+  return BasicPort(clock_port_names[0], 1); 
+}
+
+/********************************************************************
+ * Print Verilog codes to check the equivalence of output vectors 
+ *
+ * Restriction: this function only supports single clock benchmarks!
+ *******************************************************************/
+void print_verilog_testbench_check(std::fstream& fp,
+                                   const std::string& autochecked_preprocessing_flag,
+                                   const std::string& simulation_start_counter_name,
+                                   const std::string& benchmark_port_postfix,
+                                   const std::string& fpga_port_postfix,
+                                   const std::string& check_flag_port_postfix,
+                                   const std::string& error_counter_name,
+                                   const AtomContext& atom_ctx,
+                                   const VprNetlistAnnotation& netlist_annotation,
+                                   const std::vector<std::string>& clock_port_names,
+                                   const std::string& default_clock_name) {
+
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  /* Add output autocheck conditionally: only when a preprocessing flag is enable */
+  print_verilog_preprocessing_flag(fp, autochecked_preprocessing_flag); 
+
+  print_verilog_comment(fp, std::string("----- Begin checking output vectors -------"));
+
+  BasicPort clock_port = generate_verilog_testbench_clock_port(clock_port_names, default_clock_name);
+
+  print_verilog_comment(fp, std::string("----- Skip the first falling edge of clock, it is for initialization -------"));
+
+  BasicPort sim_start_port(simulation_start_counter_name, 1);
+
+  fp << "\t" << generate_verilog_port(VERILOG_PORT_REG, sim_start_port) << ";" << std::endl;
+  fp << std::endl;
+
+  fp << "\talways@(negedge " << generate_verilog_port(VERILOG_PORT_CONKT, clock_port) << ") begin" << std::endl;
+  fp << "\t\tif (1'b1 == " << generate_verilog_port(VERILOG_PORT_CONKT, sim_start_port) << ") begin" << std::endl;
+  fp << "\t\t";
+  print_verilog_register_connection(fp, sim_start_port, sim_start_port, true);
+  fp << "\t\tend else begin" << std::endl;
+
+  for (const AtomBlockId& atom_blk : atom_ctx.nlist.blocks()) {
+    /* Bypass non-I/O atom blocks ! */
+    if ( (AtomBlockType::INPAD != atom_ctx.nlist.block_type(atom_blk))
+      && (AtomBlockType::OUTPAD != atom_ctx.nlist.block_type(atom_blk)) ) {
+      continue;
+    }
+
+    /* The block may be renamed as it contains special characters which violate Verilog syntax */
+    std::string block_name = atom_ctx.nlist.block_name(atom_blk);
+    if (true == netlist_annotation.is_block_renamed(atom_blk)) {
+      block_name = netlist_annotation.block_name(atom_blk);
+    } 
+
+    if (AtomBlockType::OUTPAD == atom_ctx.nlist.block_type(atom_blk)) {
+     fp << "\t\t\tif(!(" << block_name << fpga_port_postfix;
+     fp << " === " << block_name << benchmark_port_postfix;
+     fp << ") && !(" << block_name << benchmark_port_postfix;
+     fp << " === 1'bx)) begin" << std::endl;
+     fp << "\t\t\t\t" << block_name << check_flag_port_postfix << " <= 1'b1;" << std::endl;
+     fp << "\t\t\tend else begin" << std::endl;
+     fp << "\t\t\t\t" << block_name << check_flag_port_postfix << "<= 1'b0;" << std::endl;
+     fp << "\t\t\tend" << std::endl; 
+    }
+  } 
+  fp << "\t\tend" << std::endl;
+  fp << "\tend" << std::endl;
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+
+  for (const AtomBlockId& atom_blk : atom_ctx.nlist.blocks()) {
+    /* Only care about output atom blocks ! */
+    if (AtomBlockType::OUTPAD != atom_ctx.nlist.block_type(atom_blk)) {
+      continue;
+    }
+
+    /* The block may be renamed as it contains special characters which violate Verilog syntax */
+    std::string block_name = atom_ctx.nlist.block_name(atom_blk);
+    if (true == netlist_annotation.is_block_renamed(atom_blk)) {
+      block_name = netlist_annotation.block_name(atom_blk);
+    } 
+
+    fp << "\talways@(posedge " << block_name << check_flag_port_postfix << ") begin" << std::endl;
+    fp << "\t\tif(" << block_name << check_flag_port_postfix << ") begin" << std::endl;
+    fp << "\t\t\t" << error_counter_name << " = " << error_counter_name << " + 1;" << std::endl;
+    fp << "\t\t\t$display(\"Mismatch on " << block_name << fpga_port_postfix << " at time = " << std::string("%t") << "\", $realtime);" << std::endl;
+    fp << "\t\tend" << std::endl;
+    fp << "\tend" << std::endl;
+
+    /* Add an empty line as splitter */
+    fp << std::endl;
+  }
+
+  /* Condition ends */
+  print_verilog_endif(fp);
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Generate random stimulus for the clock port
+ * This function is designed to drive the clock port of a benchmark module
+ * If there is no clock port found, we will give a default clock name
+ * In such case, this clock will not be wired to the benchmark module
+ * but be only used as a synchronizer in verification
+ *******************************************************************/
+void print_verilog_testbench_clock_stimuli(std::fstream& fp,
+                                           const SimulationSetting& simulation_parameters,
+                                           const BasicPort& clock_port) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  print_verilog_comment(fp, std::string("----- Clock Initialization -------"));
+
+  fp << "\tinitial begin" << std::endl;
+  /* Create clock stimuli */
+  fp << "\t\t" << generate_verilog_port(VERILOG_PORT_CONKT, clock_port) << " <= 1'b0;" << std::endl;
+  fp << "\t\twhile(1) begin" << std::endl;
+  fp << "\t\t\t#" << std::setprecision(10) << ((0.5/simulation_parameters.operating_clock_frequency())/VERILOG_SIM_TIMESCALE) << std::endl;
+  fp << "\t\t\t" << generate_verilog_port(VERILOG_PORT_CONKT, clock_port);
+  fp << " <= !";
+  fp << generate_verilog_port(VERILOG_PORT_CONKT, clock_port);
+  fp << ";" << std::endl;
+  fp << "\t\tend" << std::endl;
+
+  fp << "\tend" << std::endl;
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Generate random stimulus for the input ports (non-clock signals)
+ * For clock signals, please use print_verilog_testbench_clock_stimuli
+ *******************************************************************/
+void print_verilog_testbench_random_stimuli(std::fstream& fp,
+                                            const AtomContext& atom_ctx,
+                                            const VprNetlistAnnotation& netlist_annotation,
+                                            const std::vector<std::string>& clock_port_names,
+                                            const std::string& check_flag_port_postfix,
+                                            const BasicPort& clock_port) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  print_verilog_comment(fp, std::string("----- Input Initialization -------"));
+
+  fp << "\tinitial begin" << std::endl;
+
+  for (const AtomBlockId& atom_blk : atom_ctx.nlist.blocks()) {
+    /* Bypass non-I/O atom blocks ! */
+    if ( (AtomBlockType::INPAD != atom_ctx.nlist.block_type(atom_blk))
+      && (AtomBlockType::OUTPAD != atom_ctx.nlist.block_type(atom_blk)) ) {
+      continue;
+    }
+
+    /* The block may be renamed as it contains special characters which violate Verilog syntax */
+    std::string block_name = atom_ctx.nlist.block_name(atom_blk);
+    if (true == netlist_annotation.is_block_renamed(atom_blk)) {
+      block_name = netlist_annotation.block_name(atom_blk);
+    } 
+
+    /* Bypass clock ports */
+    if (clock_port_names.end() != std::find(clock_port_names.begin(), clock_port_names.end(), block_name)) {
+      continue;
+    }
+
+    /* TODO: find the clock inputs will be initialized later */
+    if (AtomBlockType::INPAD == atom_ctx.nlist.block_type(atom_blk)) {
+      fp << "\t\t" << block_name << " <= 1'b0;" << std::endl;
+    }
+  }
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+  
+  /* Set 0 to registers for checking flags */
+  for (const AtomBlockId& atom_blk : atom_ctx.nlist.blocks()) {
+    /* Bypass non-I/O atom blocks ! */
+    if (AtomBlockType::OUTPAD != atom_ctx.nlist.block_type(atom_blk)) {
+      continue;
+    }
+
+    /* The block may be renamed as it contains special characters which violate Verilog syntax */
+    std::string block_name = atom_ctx.nlist.block_name(atom_blk);
+    if (true == netlist_annotation.is_block_renamed(atom_blk)) {
+      block_name = netlist_annotation.block_name(atom_blk);
+    } 
+
+    /* Each logical block assumes a single-width port */
+    BasicPort output_port(std::string(block_name + check_flag_port_postfix), 1); 
+    fp << "\t\t" << generate_verilog_port(VERILOG_PORT_CONKT, output_port) << " <= 1'b0;" << std::endl;
+  }
+
+  fp << "\tend" << std::endl;
+  /* Finish initialization */
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+
+  // Not ready yet to determine if input is reset
+/*
+  fprintf(fp, "//----- Reset Stimulis\n");      
+  fprintf(fp, "  initial begin\n");
+  fprintf(fp, "    #%.3f\n",(rand() % 10) + 0.001);
+  fprintf(fp, "    %s <= !%s;\n", reset_input_name, reset_input_name);
+  fprintf(fp, "    #%.3f\n",(rand() % 10) + 0.001);
+  fprintf(fp, "    %s <= !%s;\n", reset_input_name, reset_input_name);
+  fprintf(fp, "    while(1) begin\n");
+  fprintf(fp, "      #%.3f\n", (rand() % 15) + 0.5);
+  fprintf(fp, "      %s <= !%s;\n", reset_input_name, reset_input_name);
+  fprintf(fp, "      #%.3f\n", (rand() % 10000) + 200);
+  fprintf(fp, "      %s <= !%s;\n", reset_input_name, reset_input_name);
+  fprintf(fp, "    end\n");
+  fprintf(fp, "  end\n\n");  
+*/
+
+  print_verilog_comment(fp, std::string("----- Input Stimulus -------"));
+  fp << "\talways@(negedge " << generate_verilog_port(VERILOG_PORT_CONKT, clock_port) << ") begin" << std::endl;
+
+  for (const AtomBlockId& atom_blk : atom_ctx.nlist.blocks()) {
+    /* Bypass non-I/O atom blocks ! */
+    if ( (AtomBlockType::INPAD != atom_ctx.nlist.block_type(atom_blk))
+      && (AtomBlockType::OUTPAD != atom_ctx.nlist.block_type(atom_blk)) ) {
+      continue;
+    }
+
+    /* The block may be renamed as it contains special characters which violate Verilog syntax */
+    std::string block_name = atom_ctx.nlist.block_name(atom_blk);
+    if (true == netlist_annotation.is_block_renamed(atom_blk)) {
+      block_name = netlist_annotation.block_name(atom_blk);
+    } 
+
+    /* Bypass clock ports */
+    if (clock_port_names.end() != std::find(clock_port_names.begin(), clock_port_names.end(), block_name)) {
+      continue;
+    }
+
+    /* TODO: find the clock inputs will be initialized later */
+    if (AtomBlockType::INPAD == atom_ctx.nlist.block_type(atom_blk)) {
+      fp << "\t\t" << block_name << " <= $random;" << std::endl;
+    }
+  }
+
+  fp << "\tend" << std::endl;
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Print Verilog declaration of shared ports appear in testbenches
+ * which are 
+ * 1. the shared input ports (registers) to drive both 
+ *    FPGA fabric and benchmark instance
+ * 2. the output ports (wires) for both FPGA fabric and benchmark instance
+ * 3. the checking flag ports to evaluate if outputs matches under the
+ *    same input vectors
+ *******************************************************************/
+void print_verilog_testbench_shared_ports(std::fstream& fp,
+                                          const AtomContext& atom_ctx,
+                                          const VprNetlistAnnotation& netlist_annotation,
+                                          const std::vector<std::string>& clock_port_names,
+                                          const std::string& benchmark_output_port_postfix,
+                                          const std::string& fpga_output_port_postfix,
+                                          const std::string& check_flag_port_postfix,
+                                          const std::string& autocheck_preprocessing_flag) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  /* Instantiate register for inputs stimulis */
+  print_verilog_comment(fp, std::string("----- Shared inputs -------"));
+  for (const AtomBlockId& atom_blk : atom_ctx.nlist.blocks()) {
+    /* We care only those logic blocks which are input I/Os */
+    if (AtomBlockType::INPAD != atom_ctx.nlist.block_type(atom_blk)) {
+      continue;
+    }
+
+    /* The block may be renamed as it contains special characters which violate Verilog syntax */
+    std::string block_name = atom_ctx.nlist.block_name(atom_blk);
+    if (true == netlist_annotation.is_block_renamed(atom_blk)) {
+      block_name = netlist_annotation.block_name(atom_blk);
+    } 
+
+    /* TODO: Skip clocks because they are handled in another function */
+    if (clock_port_names.end() != std::find(clock_port_names.begin(), clock_port_names.end(), block_name)) {
+      continue;
+    }
+   
+    /* Each logical block assumes a single-width port */
+    BasicPort input_port(block_name, 1); 
+    fp << "\t" << generate_verilog_port(VERILOG_PORT_REG, input_port) << ";" << std::endl;
+  }
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+
+  /* Instantiate wires for FPGA fabric outputs */
+  print_verilog_comment(fp, std::string("----- FPGA fabric outputs -------"));
+
+  for (const AtomBlockId& atom_blk : atom_ctx.nlist.blocks()) {
+    /* We care only those logic blocks which are output I/Os */
+    if (AtomBlockType::OUTPAD != atom_ctx.nlist.block_type(atom_blk)) {
+      continue;
+    }
+
+    /* The block may be renamed as it contains special characters which violate Verilog syntax */
+    std::string block_name = atom_ctx.nlist.block_name(atom_blk);
+    if (true == netlist_annotation.is_block_renamed(atom_blk)) {
+      block_name = netlist_annotation.block_name(atom_blk);
+    } 
+
+    /* Each logical block assumes a single-width port */
+    BasicPort output_port(std::string(block_name + fpga_output_port_postfix), 1); 
+    fp << "\t" << generate_verilog_port(VERILOG_PORT_WIRE, output_port) << ";" << std::endl;
+  }
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+
+  /* Benchmark is instanciated conditionally: only when a preprocessing flag is enable */
+  print_verilog_preprocessing_flag(fp, std::string(autocheck_preprocessing_flag)); 
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+
+  /* Instantiate wire for benchmark output */
+  print_verilog_comment(fp, std::string("----- Benchmark outputs -------"));
+  for (const AtomBlockId& atom_blk : atom_ctx.nlist.blocks()) {
+    /* We care only those logic blocks which are output I/Os */
+    if (AtomBlockType::OUTPAD != atom_ctx.nlist.block_type(atom_blk)) {
+      continue;
+    }
+
+    /* The block may be renamed as it contains special characters which violate Verilog syntax */
+    std::string block_name = atom_ctx.nlist.block_name(atom_blk);
+    if (true == netlist_annotation.is_block_renamed(atom_blk)) {
+      block_name = netlist_annotation.block_name(atom_blk);
+    } 
+
+    /* Each logical block assumes a single-width port */
+    BasicPort output_port(std::string(block_name + benchmark_output_port_postfix), 1); 
+    fp << "\t" << generate_verilog_port(VERILOG_PORT_WIRE, output_port) << ";" << std::endl;
+  }
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+
+  /* Instantiate register for output comparison */
+  print_verilog_comment(fp, std::string("----- Output vectors checking flags -------"));
+  for (const AtomBlockId& atom_blk : atom_ctx.nlist.blocks()) {
+    /* We care only those logic blocks which are output I/Os */
+    if (AtomBlockType::OUTPAD != atom_ctx.nlist.block_type(atom_blk)) {
+      continue;
+    }
+
+    /* The block may be renamed as it contains special characters which violate Verilog syntax */
+    std::string block_name = atom_ctx.nlist.block_name(atom_blk);
+    if (true == netlist_annotation.is_block_renamed(atom_blk)) {
+      block_name = netlist_annotation.block_name(atom_blk);
+    } 
+
+    /* Each logical block assumes a single-width port */
+    BasicPort output_port(std::string(block_name + check_flag_port_postfix), 1); 
+    fp << "\t" << generate_verilog_port(VERILOG_PORT_REG, output_port) << ";" << std::endl;
+  }
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+
+  /* Condition ends for the benchmark instanciation */
+  print_verilog_endif(fp);
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+}
+
+} /* end namespace openfpga */

openfpga/src/fpga_verilog/verilog_testbench_utils.h

@@ -0,0 +1,94 @@
+#ifndef VERILOG_TESTBENCH_UTILS_H
+#define VERILOG_TESTBENCH_UTILS_H
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+#include <fstream>
+#include <string>
+#include <vector>
+#include "module_manager.h"
+#include "vpr_context.h"
+#include "io_location_map.h"
+#include "vpr_netlist_annotation.h"
+#include "simulation_setting.h"
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+void print_verilog_testbench_fpga_instance(std::fstream& fp,
+                                           const ModuleManager& module_manager,
+                                           const ModuleId& top_module,
+                                           const std::string& top_instance_name);
+
+void print_verilog_testbench_benchmark_instance(std::fstream& fp,
+                                                const std::string& module_name,
+                                                const std::string& instance_name,
+                                                const std::string& module_input_port_postfix,
+                                                const std::string& module_output_port_postfix,
+                                                const std::string& output_port_postfix,
+                                                const AtomContext& atom_ctx,
+                                                const VprNetlistAnnotation& netlist_annotation,
+                                                const bool& use_explicit_port_map);
+
+void print_verilog_testbench_connect_fpga_ios(std::fstream& fp,
+                                              const ModuleManager& module_manager,
+                                              const ModuleId& top_module,
+                                              const AtomContext& atom_ctx,
+                                              const PlacementContext& place_ctx,
+                                              const IoLocationMap& io_location_map,
+                                              const VprNetlistAnnotation& netlist_annotation,
+                                              const std::string& io_input_port_name_postfix,
+                                              const std::string& io_output_port_name_postfix,
+                                              const size_t& unused_io_value);
+
+void print_verilog_timeout_and_vcd(std::fstream& fp,
+                                   const std::string& icarus_preprocessing_flag,
+                                   const std::string& module_name,
+                                   const std::string& vcd_fname,
+                                   const std::string& simulation_start_counter_name,
+                                   const std::string& error_counter_name,
+                                   const int& simulation_time);
+
+BasicPort generate_verilog_testbench_clock_port(const std::vector<std::string>& clock_port_names,
+                                                const std::string& default_clock_name);
+
+void print_verilog_testbench_check(std::fstream& fp,
+                                   const std::string& autochecked_preprocessing_flag,
+                                   const std::string& simulation_start_counter_name,
+                                   const std::string& benchmark_port_postfix,
+                                   const std::string& fpga_port_postfix,
+                                   const std::string& check_flag_port_postfix,
+                                   const std::string& error_counter_name,
+                                   const AtomContext& atom_ctx,
+                                   const VprNetlistAnnotation& netlist_annotation,
+                                   const std::vector<std::string>& clock_port_names,
+                                   const std::string& default_clock_name);
+
+void print_verilog_testbench_clock_stimuli(std::fstream& fp,
+                                           const SimulationSetting& simulation_parameters,
+                                           const BasicPort& clock_port);
+
+void print_verilog_testbench_random_stimuli(std::fstream& fp,
+                                            const AtomContext& atom_ctx,
+                                            const VprNetlistAnnotation& netlist_annotation,
+                                            const std::vector<std::string>& clock_port_names,
+                                            const std::string& check_flag_port_postfix,
+                                            const BasicPort& clock_port);
+
+void print_verilog_testbench_shared_ports(std::fstream& fp,
+                                          const AtomContext& atom_ctx,
+                                          const VprNetlistAnnotation& netlist_annotation,
+                                          const std::vector<std::string>& clock_port_names,
+                                          const std::string& benchmark_output_port_postfix,
+                                          const std::string& fpga_output_port_postfix,
+                                          const std::string& check_flag_port_postfix,
+                                          const std::string& autocheck_preprocessing_flag);
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/fpga_verilog/verilog_top_testbench.cpp

@@ -0,0 +1,900 @@
+/********************************************************************
+ * This file includes functions that are used to create
+ * an auto-check top-level testbench for a FPGA fabric
+ *******************************************************************/
+#include <fstream>
+#include <iomanip>
+#include <algorithm>
+
+
+/* Headers from vtrutil library */
+#include "vtr_log.h"
+#include "vtr_assert.h"
+#include "vtr_time.h"
+
+/* Headers from openfpgautil library */
+#include "openfpga_port.h"
+#include "openfpga_digest.h"
+
+#include "bitstream_manager_utils.h"
+
+#include "openfpga_naming.h"
+#include "simulation_utils.h"
+#include "openfpga_atom_netlist_utils.h"
+
+#include "verilog_constants.h"
+#include "verilog_writer_utils.h"
+#include "verilog_testbench_utils.h"
+#include "verilog_top_testbench.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/********************************************************************
+ * Local variables used only in this file
+ *******************************************************************/
+constexpr char* TOP_TESTBENCH_REFERENCE_INSTANCE_NAME = "REF_DUT";
+constexpr char* TOP_TESTBENCH_FPGA_INSTANCE_NAME = "FPGA_DUT";
+constexpr char* TOP_TESTBENCH_REFERENCE_OUTPUT_POSTFIX = "_benchmark";
+constexpr char* TOP_TESTBENCH_FPGA_OUTPUT_POSTFIX = "_fpga";
+
+constexpr char* TOP_TESTBENCH_CHECKFLAG_PORT_POSTFIX = "_flag";
+
+constexpr char* TOP_TESTBENCH_CC_PROG_TASK_NAME = "prog_cycle_task";
+
+constexpr char* TOP_TESTBENCH_SIM_START_PORT_NAME = "sim_start";
+
+constexpr int TOP_TESTBENCH_MAGIC_NUMBER_FOR_SIMULATION_TIME = 200;
+constexpr char* TOP_TESTBENCH_ERROR_COUNTER = "nb_error";
+
+constexpr char* TOP_TB_RESET_PORT_NAME = "greset";
+constexpr char* TOP_TB_SET_PORT_NAME = "gset";
+constexpr char* TOP_TB_PROG_RESET_PORT_NAME = "prog_reset";
+constexpr char* TOP_TB_PROG_SET_PORT_NAME = "prog_set";
+constexpr char* TOP_TB_CONFIG_DONE_PORT_NAME = "config_done";
+constexpr char* TOP_TB_OP_CLOCK_PORT_NAME = "op_clock";
+constexpr char* TOP_TB_PROG_CLOCK_PORT_NAME = "prog_clock";
+constexpr char* TOP_TB_INOUT_REG_POSTFIX = "_reg";
+constexpr char* TOP_TB_CLOCK_REG_POSTFIX = "_reg";
+
+constexpr char* AUTOCHECK_TOP_TESTBENCH_VERILOG_MODULE_POSTFIX = "_autocheck_top_tb";
+
+/********************************************************************
+ * Print local wires for configuration chain protocols
+ *******************************************************************/
+static 
+void print_verilog_top_testbench_config_chain_port(std::fstream& fp) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  /* Print the head of configuraion-chains here */
+  print_verilog_comment(fp, std::string("---- Configuration-chain head -----"));
+  BasicPort config_chain_head_port(generate_configuration_chain_head_name(), 1);
+  fp << generate_verilog_port(VERILOG_PORT_REG, config_chain_head_port) << ";" << std::endl;
+
+  /* Print the tail of configuration-chains here */
+  print_verilog_comment(fp, std::string("---- Configuration-chain tail -----"));
+  BasicPort config_chain_tail_port(generate_configuration_chain_tail_name(), 1);
+  fp << generate_verilog_port(VERILOG_PORT_WIRE, config_chain_tail_port) << ";" << std::endl;
+}
+
+/********************************************************************
+ * Print local wires for different types of configuration protocols
+ *******************************************************************/
+static 
+void print_verilog_top_testbench_config_protocol_port(std::fstream& fp,
+                                                      const e_config_protocol_type& sram_orgz_type) {
+  switch(sram_orgz_type) {
+  case CONFIG_MEM_STANDALONE:
+    /* TODO */
+    break;
+  case CONFIG_MEM_SCAN_CHAIN:
+    print_verilog_top_testbench_config_chain_port(fp);
+    break;
+  case CONFIG_MEM_MEMORY_BANK:
+    /* TODO */
+    break;
+  default:
+    VTR_LOGF_ERROR(__FILE__, __LINE__,
+                   "Invalid type of SRAM organization!\n");
+    exit(1);
+  }
+}
+
+/********************************************************************
+ * Wire the global ports of FPGA fabric to local wires
+ *******************************************************************/
+static 
+void print_verilog_top_testbench_global_ports_stimuli(std::fstream& fp,
+                                                      const ModuleManager& module_manager,
+                                                      const ModuleId& top_module,
+                                                      const CircuitLibrary& circuit_lib,
+                                                      const std::vector<CircuitPortId>& global_ports) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  print_verilog_comment(fp, std::string("----- Begin connecting global ports of FPGA fabric to stimuli -----"));
+
+  /* Connect global clock ports to operating or programming clock signal */
+  for (const CircuitPortId& model_global_port : global_ports) {
+    if (CIRCUIT_MODEL_PORT_CLOCK != circuit_lib.port_type(model_global_port)) {
+      continue;
+    }
+    /* Reach here, it means we have a global clock to deal with:
+     * 1. if the port is identified as a programming clock, 
+     *    connect it to the local wire of programming clock
+     * 2. if the port is identified as an operating clock 
+     *    connect it to the local wire of operating clock
+     */
+    /* Find the module port */
+    ModulePortId module_global_port = module_manager.find_module_port(top_module, circuit_lib.port_prefix(model_global_port));
+    VTR_ASSERT(true == module_manager.valid_module_port_id(top_module, module_global_port));
+
+    BasicPort stimuli_clock_port;
+    if (true == circuit_lib.port_is_prog(model_global_port)) {
+      stimuli_clock_port.set_name(std::string(TOP_TB_PROG_CLOCK_PORT_NAME));
+      stimuli_clock_port.set_width(1);
+    } else {
+      VTR_ASSERT_SAFE(false == circuit_lib.port_is_prog(model_global_port));
+      stimuli_clock_port.set_name(std::string(TOP_TB_OP_CLOCK_PORT_NAME));
+      stimuli_clock_port.set_width(1);
+    }
+    /* Wire the port to the input stimuli:
+     * The wiring will be inverted if the default value of the global port is 1
+     * Otherwise, the wiring will not be inverted!
+     */
+    print_verilog_wire_connection(fp, module_manager.module_port(top_module, module_global_port), 
+                                  stimuli_clock_port, 
+                                  1 == circuit_lib.port_default_value(model_global_port)); 
+  }
+
+  /* Connect global configuration done ports to configuration done signal */
+  for (const CircuitPortId& model_global_port : global_ports) {
+    /* Bypass clock signals, they have been processed */
+    if (CIRCUIT_MODEL_PORT_CLOCK == circuit_lib.port_type(model_global_port)) {
+      continue;
+    }
+    if (false == circuit_lib.port_is_config_enable(model_global_port)) {
+      continue;
+    }
+    /* Reach here, it means we have a configuration done port to deal with */
+    /* Find the module port */
+    ModulePortId module_global_port = module_manager.find_module_port(top_module, circuit_lib.port_prefix(model_global_port));
+    VTR_ASSERT(true == module_manager.valid_module_port_id(top_module, module_global_port));
+
+    BasicPort stimuli_config_done_port(std::string(TOP_TB_CONFIG_DONE_PORT_NAME), 1);
+    /* Wire the port to the input stimuli:
+     * The wiring will be inverted if the default value of the global port is 1
+     * Otherwise, the wiring will not be inverted!
+     */
+    print_verilog_wire_connection(fp, module_manager.module_port(top_module, module_global_port), 
+                                  stimuli_config_done_port, 
+                                  1 == circuit_lib.port_default_value(model_global_port)); 
+  }
+
+  /* Connect global reset ports to operating or programming reset signal */
+  for (const CircuitPortId& model_global_port : global_ports) {
+    /* Bypass clock signals, they have been processed */
+    if (CIRCUIT_MODEL_PORT_CLOCK == circuit_lib.port_type(model_global_port)) {
+      continue;
+    }
+    /* Bypass config_done signals, they have been processed */
+    if (true == circuit_lib.port_is_config_enable(model_global_port)) {
+      continue;
+    }
+
+    if (false == circuit_lib.port_is_reset(model_global_port)) {
+      continue;
+    }
+    /* Reach here, it means we have a reset port to deal with */
+    /* Find the module port */
+    ModulePortId module_global_port = module_manager.find_module_port(top_module, circuit_lib.port_prefix(model_global_port));
+    VTR_ASSERT(true == module_manager.valid_module_port_id(top_module, module_global_port));
+
+    BasicPort stimuli_reset_port;
+    if (true == circuit_lib.port_is_prog(model_global_port)) {
+      stimuli_reset_port.set_name(std::string(TOP_TB_PROG_RESET_PORT_NAME));
+      stimuli_reset_port.set_width(1);
+    } else {
+      VTR_ASSERT_SAFE(false == circuit_lib.port_is_prog(model_global_port));
+      stimuli_reset_port.set_name(std::string(TOP_TB_RESET_PORT_NAME));
+      stimuli_reset_port.set_width(1);
+    }
+    /* Wire the port to the input stimuli:
+     * The wiring will be inverted if the default value of the global port is 1
+     * Otherwise, the wiring will not be inverted!
+     */
+    print_verilog_wire_connection(fp, module_manager.module_port(top_module, module_global_port), 
+                                  stimuli_reset_port, 
+                                  1 == circuit_lib.port_default_value(model_global_port)); 
+  }
+
+  /* Connect global set ports to operating or programming set signal */
+  for (const CircuitPortId& model_global_port : global_ports) {
+    /* Bypass clock signals, they have been processed */
+    if (CIRCUIT_MODEL_PORT_CLOCK == circuit_lib.port_type(model_global_port)) {
+      continue;
+    }
+    /* Bypass config_done signals, they have been processed */
+    if (true == circuit_lib.port_is_config_enable(model_global_port)) {
+      continue;
+    }
+
+    /* Bypass reset signals, they have been processed */
+    if (true == circuit_lib.port_is_reset(model_global_port)) {
+      continue;
+    }
+
+    if (false == circuit_lib.port_is_set(model_global_port)) {
+      continue;
+    }
+    /* Reach here, it means we have a set port to deal with */
+    /* Find the module port */
+    ModulePortId module_global_port = module_manager.find_module_port(top_module, circuit_lib.port_prefix(model_global_port));
+    VTR_ASSERT(true == module_manager.valid_module_port_id(top_module, module_global_port));
+
+    BasicPort stimuli_set_port;
+    if (true == circuit_lib.port_is_prog(model_global_port)) {
+      stimuli_set_port.set_name(std::string(TOP_TB_PROG_SET_PORT_NAME));
+      stimuli_set_port.set_width(1);
+    } else {
+      VTR_ASSERT_SAFE(false == circuit_lib.port_is_prog(model_global_port));
+      stimuli_set_port.set_name(std::string(TOP_TB_SET_PORT_NAME));
+      stimuli_set_port.set_width(1);
+    }
+    /* Wire the port to the input stimuli:
+     * The wiring will be inverted if the default value of the global port is 1
+     * Otherwise, the wiring will not be inverted!
+     */
+    print_verilog_wire_connection(fp, module_manager.module_port(top_module, module_global_port), 
+                                  stimuli_set_port, 
+                                  1 == circuit_lib.port_default_value(model_global_port)); 
+  }
+
+  /* For the rest of global ports, wire them to constant signals */
+  for (const CircuitPortId& model_global_port : global_ports) {
+    /* Bypass clock signals, they have been processed */
+    if (CIRCUIT_MODEL_PORT_CLOCK == circuit_lib.port_type(model_global_port)) {
+      continue;
+    }
+    /* Bypass config_done signals, they have been processed */
+    if (true == circuit_lib.port_is_config_enable(model_global_port)) {
+      continue;
+    }
+
+    /* Bypass reset signals, they have been processed */
+    if (true == circuit_lib.port_is_reset(model_global_port)) {
+      continue;
+    }
+
+    /* Bypass set signals, they have been processed */
+    if (true == circuit_lib.port_is_set(model_global_port)) {
+      continue;
+    }
+
+    /* Reach here, it means we have a port to deal with */
+    /* Find the module port and wire it to constant values */
+    ModulePortId module_global_port = module_manager.find_module_port(top_module, circuit_lib.port_prefix(model_global_port));
+    VTR_ASSERT(true == module_manager.valid_module_port_id(top_module, module_global_port));
+
+    BasicPort module_port = module_manager.module_port(top_module, module_global_port);
+    std::vector<size_t> default_values(module_port.get_width(), circuit_lib.port_default_value(model_global_port));
+    print_verilog_wire_constant_values(fp, module_port, default_values);
+  }
+
+  print_verilog_comment(fp, std::string("----- End connecting global ports of FPGA fabric to stimuli -----"));
+}
+
+/********************************************************************
+ * This function prints the top testbench module declaration 
+ * and internal wires/port declaration
+ * Ports can be classified in two categories:
+ * 1. General-purpose ports, which are datapath I/Os, clock signals
+ *    for the FPGA fabric and input benchmark
+ * 2. Fabric-featured ports, which are required by configuration
+ *    protocols.
+ *    Due the difference in configuration protocols, the internal
+ *    wires and ports will be different:
+ *    (a) configuration-chain: we will have two ports,
+ *        a head and a tail for the configuration chain,
+ *        in addition to the regular ports.
+ *    (b) memory-decoders: we will have a few ports to drive 
+ *        address lines for decoders and a bit input port to feed
+ *        configuration bits
+ *******************************************************************/
+static 
+void print_verilog_top_testbench_ports(std::fstream& fp,
+                                       const ModuleManager& module_manager,
+                                       const ModuleId& top_module,
+                                       const AtomContext& atom_ctx,
+                                       const VprNetlistAnnotation& netlist_annotation,
+                                       const std::vector<std::string>& clock_port_names,
+                                       const e_config_protocol_type& sram_orgz_type,
+                                       const std::string& circuit_name){
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  /* Print module definition */
+  fp << "module " << circuit_name << std::string(AUTOCHECK_TOP_TESTBENCH_VERILOG_MODULE_POSTFIX);
+  fp << ";" << std::endl;
+
+  /* Print regular local wires:
+   * 1. global ports, i.e., reset, set and clock signals
+   * 2. datapath I/O signals
+   */
+  /* Global ports of top-level module  */
+  print_verilog_comment(fp, std::string("----- Local wires for global ports of FPGA fabric -----"));
+  for (const BasicPort& module_port : module_manager.module_ports_by_type(top_module, ModuleManager::MODULE_GLOBAL_PORT)) {
+    fp << generate_verilog_port(VERILOG_PORT_WIRE, module_port) << ";" << std::endl;
+  }
+  /* Add an empty line as a splitter */
+  fp << std::endl;
+
+  /* Datapath I/Os of top-level module  */
+  print_verilog_comment(fp, std::string("----- Local wires for I/Os of FPGA fabric -----"));
+  for (const BasicPort& module_port : module_manager.module_ports_by_type(top_module, ModuleManager::MODULE_GPIO_PORT)) {
+    fp << generate_verilog_port(VERILOG_PORT_WIRE, module_port) << ";" << std::endl;
+  }
+  /* Add an empty line as a splitter */
+  fp << std::endl;
+
+  /* Add local wires/registers that drive stimulus
+   * We create these general purpose ports here,
+   * and then wire them to the ports of FPGA fabric depending on their usage 
+   */
+  /* Configuration done port */
+  BasicPort config_done_port(std::string(TOP_TB_CONFIG_DONE_PORT_NAME), 1);
+  fp << generate_verilog_port(VERILOG_PORT_REG, config_done_port) << ";" << std::endl;
+
+  /* Programming clock */
+  BasicPort prog_clock_port(std::string(TOP_TB_PROG_CLOCK_PORT_NAME), 1);
+  fp << generate_verilog_port(VERILOG_PORT_WIRE, prog_clock_port) << ";" << std::endl;
+  BasicPort prog_clock_register_port(std::string(std::string(TOP_TB_PROG_CLOCK_PORT_NAME) + std::string(TOP_TB_CLOCK_REG_POSTFIX)), 1);
+  fp << generate_verilog_port(VERILOG_PORT_REG, prog_clock_register_port) << ";" << std::endl;
+
+  /* Operating clock */
+  BasicPort op_clock_port(std::string(TOP_TB_OP_CLOCK_PORT_NAME), 1);
+  fp << generate_verilog_port(VERILOG_PORT_WIRE, op_clock_port) << ";" << std::endl;
+  BasicPort op_clock_register_port(std::string(std::string(TOP_TB_OP_CLOCK_PORT_NAME) + std::string(TOP_TB_CLOCK_REG_POSTFIX)), 1);
+  fp << generate_verilog_port(VERILOG_PORT_REG, op_clock_register_port) << ";" << std::endl;
+
+  /* Programming set and reset */
+  BasicPort prog_reset_port(std::string(TOP_TB_PROG_RESET_PORT_NAME), 1);
+  fp << generate_verilog_port(VERILOG_PORT_REG, prog_reset_port) << ";" << std::endl;
+  BasicPort prog_set_port(std::string(TOP_TB_PROG_SET_PORT_NAME), 1);
+  fp << generate_verilog_port(VERILOG_PORT_REG, prog_set_port) << ";" << std::endl;
+
+  /* Global set and reset */
+  BasicPort reset_port(std::string(TOP_TB_RESET_PORT_NAME), 1);
+  fp << generate_verilog_port(VERILOG_PORT_REG, reset_port) << ";" << std::endl;
+  BasicPort set_port(std::string(TOP_TB_SET_PORT_NAME), 1);
+  fp << generate_verilog_port(VERILOG_PORT_REG, set_port) << ";" << std::endl;
+
+  /* Configuration ports depend on the organization of SRAMs */
+  print_verilog_top_testbench_config_protocol_port(fp, sram_orgz_type);
+
+  /* Create a clock port if the benchmark have one but not in the default name! 
+   * We will wire the clock directly to the operating clock directly
+   */
+  for (const std::string clock_port_name : clock_port_names) {
+    if (0 == clock_port_name.compare(op_clock_port.get_name())) {
+      continue;
+    }
+    /* Ensure the clock port name is not a duplication of global ports of the FPGA module */
+    bool print_clock_port = true;
+    for (const BasicPort& module_port : module_manager.module_ports_by_type(top_module, ModuleManager::MODULE_GLOBAL_PORT)) {
+      if (0 == clock_port_name.compare(module_port.get_name())) {
+        print_clock_port = false;
+      }
+    }
+    if (false == print_clock_port) {
+      continue;
+    }
+
+    /* Print the clock and wire it to op_clock */
+    print_verilog_comment(fp, std::string("----- Create a clock for benchmark and wire it to op_clock -------"));
+    BasicPort clock_port(clock_port_name, 1);
+    fp << "\t" << generate_verilog_port(VERILOG_PORT_WIRE, clock_port) << ";" << std::endl;
+    print_verilog_wire_connection(fp, clock_port, op_clock_port, false);
+  }
+
+  print_verilog_testbench_shared_ports(fp, atom_ctx, netlist_annotation,
+                                       clock_port_names,
+                                       std::string(TOP_TESTBENCH_REFERENCE_OUTPUT_POSTFIX),
+                                       std::string(TOP_TESTBENCH_FPGA_OUTPUT_POSTFIX),
+                                       std::string(TOP_TESTBENCH_CHECKFLAG_PORT_POSTFIX),
+                                       std::string(AUTOCHECKED_SIMULATION_FLAG));
+
+  /* Instantiate an integer to count the number of error and 
+   * determine if the simulation succeed or failed
+   */
+  print_verilog_comment(fp, std::string("----- Error counter -----"));
+  fp << "\tinteger " << TOP_TESTBENCH_ERROR_COUNTER << "= 0;" << std::endl;
+}
+
+/********************************************************************
+ * Instanciate the input benchmark module
+ *******************************************************************/
+static
+void print_verilog_top_testbench_benchmark_instance(std::fstream& fp, 
+                                                    const std::string& reference_verilog_top_name,
+                                                    const AtomContext& atom_ctx,
+                                                    const VprNetlistAnnotation& netlist_annotation) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  /* Benchmark is instanciated conditionally: only when a preprocessing flag is enable */
+  print_verilog_preprocessing_flag(fp, std::string(AUTOCHECKED_SIMULATION_FLAG)); 
+
+  print_verilog_comment(fp, std::string("----- Reference Benchmark Instanication -------"));
+
+  /* Do NOT use explicit port mapping here: 
+   * VPR added a prefix of "out_" to the output ports of input benchmark
+   */
+  print_verilog_testbench_benchmark_instance(fp, reference_verilog_top_name,
+                                             std::string(TOP_TESTBENCH_REFERENCE_INSTANCE_NAME),
+                                             std::string(),
+                                             std::string(),
+                                             std::string(TOP_TESTBENCH_REFERENCE_OUTPUT_POSTFIX),
+                                             atom_ctx, netlist_annotation,
+                                             false);
+
+  print_verilog_comment(fp, std::string("----- End reference Benchmark Instanication -------"));
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+
+  /* Condition ends for the benchmark instanciation */
+  print_verilog_endif(fp);
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Print tasks (processes) in Verilog format, 
+ * which is very useful in generating stimuli for each clock cycle 
+ * This function is tuned for configuration-chain manipulation: 
+ * During each programming cycle, we feed the input of scan chain with a memory bit
+ *******************************************************************/
+static 
+void print_verilog_top_testbench_load_bitstream_task_configuration_chain(std::fstream& fp) {
+
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  BasicPort prog_clock_port(std::string(TOP_TB_PROG_CLOCK_PORT_NAME), 1);
+  BasicPort cc_head_port(generate_configuration_chain_head_name(), 1);
+  BasicPort cc_head_value(generate_configuration_chain_head_name() + std::string("_val"), 1);
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+
+  /* Feed the scan-chain input at each falling edge of programming clock 
+   * It aims at avoid racing the programming clock (scan-chain data changes at the rising edge). 
+   */
+  print_verilog_comment(fp, std::string("----- Task: input values during a programming clock cycle -----"));
+  fp << "task " << std::string(TOP_TESTBENCH_CC_PROG_TASK_NAME) << ";" << std::endl;
+  fp << generate_verilog_port(VERILOG_PORT_INPUT, cc_head_value) << ";" << std::endl;
+  fp << "\tbegin" << std::endl;
+  fp << "\t\t@(negedge " << generate_verilog_port(VERILOG_PORT_CONKT, prog_clock_port) << ");" << std::endl;
+  fp << "\t\t\t"; 
+  fp << generate_verilog_port(VERILOG_PORT_CONKT, cc_head_port);
+  fp << " = ";
+  fp << generate_verilog_port(VERILOG_PORT_CONKT, cc_head_value);
+  fp << ";" << std::endl;
+
+  fp << "\tend" << std::endl;
+  fp << "endtask" << std::endl;
+
+  /* Add an empty line as splitter */
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Print tasks, which is very useful in generating stimuli for each clock cycle 
+ *******************************************************************/
+static 
+void print_verilog_top_testbench_load_bitstream_task(std::fstream& fp,
+                                                     const e_config_protocol_type& sram_orgz_type) {
+  switch (sram_orgz_type) {
+  case CONFIG_MEM_STANDALONE:
+    break;
+  case CONFIG_MEM_SCAN_CHAIN:
+    print_verilog_top_testbench_load_bitstream_task_configuration_chain(fp);
+    break;
+  case CONFIG_MEM_MEMORY_BANK:
+    /* TODO: 
+    dump_verilog_top_testbench_stimuli_serial_version_tasks_memory_bank(cur_sram_orgz_info, fp);
+     */
+    break;
+  default:
+    VTR_LOGF_ERROR(__FILE__, __LINE__,
+                   "Invalid type of SRAM organization!\n");
+    exit(1);
+  }
+}
+
+/********************************************************************
+ * Print generatic input stimuli for the top testbench
+ * include:
+ * 1. configuration done signal
+ * 2. programming clock
+ * 3. operating clock
+ * 4. programming reset signal
+ * 5. programming set signal
+ * 6. reset signal
+ * 7. set signal
+ *******************************************************************/
+static 
+void print_verilog_top_testbench_generic_stimulus(std::fstream& fp,
+                                                  const size_t& num_config_clock_cycles,
+                                                  const float& prog_clock_period,
+                                                  const float& op_clock_period,
+                                                  const float& timescale) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  print_verilog_comment(fp, std::string("----- Number of clock cycles in configuration phase: " + std::to_string(num_config_clock_cycles) + " -----")); 
+
+  BasicPort config_done_port(std::string(TOP_TB_CONFIG_DONE_PORT_NAME), 1);
+
+  BasicPort op_clock_port(std::string(TOP_TB_OP_CLOCK_PORT_NAME), 1);
+  BasicPort op_clock_register_port(std::string(std::string(TOP_TB_OP_CLOCK_PORT_NAME) + std::string(TOP_TB_CLOCK_REG_POSTFIX)), 1);
+
+  BasicPort prog_clock_port(std::string(TOP_TB_PROG_CLOCK_PORT_NAME), 1);
+  BasicPort prog_clock_register_port(std::string(std::string(TOP_TB_PROG_CLOCK_PORT_NAME) + std::string(TOP_TB_CLOCK_REG_POSTFIX)), 1);
+
+  BasicPort prog_reset_port(std::string(TOP_TB_PROG_RESET_PORT_NAME), 1);
+  BasicPort prog_set_port(std::string(TOP_TB_PROG_SET_PORT_NAME), 1);
+
+  BasicPort reset_port(std::string(TOP_TB_RESET_PORT_NAME), 1);
+  BasicPort set_port(std::string(TOP_TB_SET_PORT_NAME), 1);
+
+  /* Generate stimuli waveform for configuration done signals */
+  print_verilog_comment(fp, "----- Begin configuration done signal generation -----"); 
+  print_verilog_pulse_stimuli(fp, config_done_port, 
+                              0, /* Initial value */
+                              num_config_clock_cycles * prog_clock_period / timescale, 0); 
+  print_verilog_comment(fp, "----- End configuration done signal generation -----"); 
+  fp << std::endl;
+
+  /* Generate stimuli waveform for programming clock signals */
+  print_verilog_comment(fp, "----- Begin raw programming clock signal generation -----"); 
+  print_verilog_clock_stimuli(fp, prog_clock_register_port, 
+                              0, /* Initial value */
+                              0.5 * prog_clock_period / timescale,
+                              std::string()); 
+  print_verilog_comment(fp, "----- End raw programming clock signal generation -----"); 
+  fp << std::endl;
+
+  /* Programming clock should be only enabled during programming phase.
+   * When configuration is done (config_done is enabled), programming clock should be always zero.
+   */
+  print_verilog_comment(fp, std::string("----- Actual programming clock is triggered only when " + config_done_port.get_name() + " and " + prog_reset_port.get_name() + " are disabled -----")); 
+  fp << "\tassign " << generate_verilog_port(VERILOG_PORT_CONKT, prog_clock_port);
+  fp << " = " << generate_verilog_port(VERILOG_PORT_CONKT, prog_clock_register_port);
+  fp << " & (~" << generate_verilog_port(VERILOG_PORT_CONKT, config_done_port) << ")";
+  fp << " & (~" << generate_verilog_port(VERILOG_PORT_CONKT, prog_reset_port) << ")";
+  fp << ";" << std::endl;
+
+  fp << std::endl;
+
+  /* Generate stimuli waveform for operating clock signals */
+  print_verilog_comment(fp, "----- Begin raw operating clock signal generation -----"); 
+  print_verilog_clock_stimuli(fp, op_clock_register_port, 
+                              0, /* Initial value */
+                              0.5 * op_clock_period / timescale,
+                              std::string("~" + reset_port.get_name())); 
+  print_verilog_comment(fp, "----- End raw operating clock signal generation -----"); 
+
+  /* Operation clock should be enabled after programming phase finishes.
+   * Before configuration is done (config_done is enabled), operation clock should be always zero.
+   */
+  print_verilog_comment(fp, std::string("----- Actual operating clock is triggered only when " + config_done_port.get_name() + " is enabled -----")); 
+  fp << "\tassign " << generate_verilog_port(VERILOG_PORT_CONKT, op_clock_port);
+  fp << " = " << generate_verilog_port(VERILOG_PORT_CONKT, op_clock_register_port);
+  fp << " & " << generate_verilog_port(VERILOG_PORT_CONKT, config_done_port);
+  fp << ";" << std::endl;
+
+  fp << std::endl;
+
+  /* Reset signal for configuration circuit: 
+   * only enable during the first clock cycle in programming phase 
+   */
+  print_verilog_comment(fp, "----- Begin programming reset signal generation -----"); 
+  print_verilog_pulse_stimuli(fp, prog_reset_port, 
+                              1, /* Initial value */
+                              prog_clock_period / timescale, 0); 
+  print_verilog_comment(fp, "----- End programming reset signal generation -----"); 
+
+  fp << std::endl;
+
+  /* Programming set signal for configuration circuit : always disabled */
+  print_verilog_comment(fp, "----- Begin programming set signal generation: always disabled -----"); 
+  print_verilog_pulse_stimuli(fp, prog_set_port, 
+                              0, /* Initial value */
+                              prog_clock_period / timescale, 0); 
+  print_verilog_comment(fp, "----- End programming set signal generation: always disabled -----"); 
+
+  fp << std::endl;
+
+  /* Operating reset signals: only enabled during the first clock cycle in operation phase */
+  std::vector<float> reset_pulse_widths;
+  reset_pulse_widths.push_back(op_clock_period / timescale);
+  reset_pulse_widths.push_back(2 * op_clock_period / timescale);
+
+  std::vector<size_t> reset_flip_values;
+  reset_flip_values.push_back(1);
+  reset_flip_values.push_back(0);
+
+  print_verilog_comment(fp, "----- Begin operating reset signal generation -----"); 
+  print_verilog_comment(fp, "----- Reset signal is enabled until the first clock cycle in operation phase -----");
+  print_verilog_pulse_stimuli(fp, reset_port,
+                              1,
+                              reset_pulse_widths,
+                              reset_flip_values,
+                              config_done_port.get_name());
+  print_verilog_comment(fp, "----- End operating reset signal generation -----"); 
+
+  /* Operating set signal for configuration circuit : always disabled */
+  print_verilog_comment(fp, "----- Begin operating set signal generation: always disabled -----"); 
+  print_verilog_pulse_stimuli(fp, set_port, 
+                              0, /* Initial value */
+                              op_clock_period / timescale, 0); 
+  print_verilog_comment(fp, "----- End operating set signal generation: always disabled -----"); 
+
+  fp << std::endl;
+}
+
+/********************************************************************
+ * Print stimulus for a FPGA fabric with a configuration chain protocol 
+ * where configuration bits are programming in serial (one by one)
+ * Task list:
+ * 1. For clock signal, we should create voltage waveforms for two types of clock signals:
+ *    a. operation clock
+ *    b. programming clock 
+ * 2. For Set/Reset, we reset the chip after programming phase ends 
+ *    and before operation phase starts
+ * 3. For input/output clb nets (mapped to I/O grids), 
+ *    we should create voltage waveforms only after programming phase 
+ *******************************************************************/
+static 
+void print_verilog_top_testbench_configuration_chain_bitstream(std::fstream& fp,
+                                                               const BitstreamManager& bitstream_manager,
+                                                               const std::vector<ConfigBitId>& fabric_bitstream) {
+  /* Validate the file stream */
+  valid_file_stream(fp);
+
+  /* Initial value should be the first configuration bits
+   * In the rest of programming cycles, 
+   * configuration bits are fed at the falling edge of programming clock.
+   * We do not care the value of scan_chain head during the first programming cycle 
+   * It is reset anyway
+   */
+  BasicPort config_chain_head_port(generate_configuration_chain_head_name(), 1);
+  std::vector<size_t> initial_values(config_chain_head_port.get_width(), 0);
+
+  print_verilog_comment(fp, "----- Begin bitstream loading during configuration phase -----");
+  fp << "initial" << std::endl;
+  fp << "\tbegin" << std::endl;
+  print_verilog_comment(fp, "----- Configuration chain default input -----");
+  fp << "\t\t";
+  fp << generate_verilog_port_constant_values(config_chain_head_port, initial_values);
+  fp << ";";
+
+  fp << std::endl;
+
+  /* Attention: the configuration chain protcol requires the last configuration bit is fed first
+   * We will visit the fabric bitstream in a reverse way  
+   */
+  std::vector<ConfigBitId> cc_bitstream = fabric_bitstream;
+  std::reverse(cc_bitstream.begin(), cc_bitstream.end());
+  for (const ConfigBitId& bit_id : cc_bitstream) {
+    fp << "\t\t" << std::string(TOP_TESTBENCH_CC_PROG_TASK_NAME);
+    fp << "(1'b" << (size_t)bitstream_manager.bit_value(bit_id) << ");" << std::endl;
+  }
+
+  /* Raise the flag of configuration done when bitstream loading is complete */
+  BasicPort prog_clock_port(std::string(TOP_TB_PROG_CLOCK_PORT_NAME), 1);
+  fp << "\t\t@(negedge " << generate_verilog_port(VERILOG_PORT_CONKT, prog_clock_port) << ");" << std::endl;
+  
+  BasicPort config_done_port(std::string(TOP_TB_CONFIG_DONE_PORT_NAME), 1);
+  fp << "\t\t\t";
+  fp << generate_verilog_port(VERILOG_PORT_CONKT, config_done_port);
+  fp << " <= ";
+  std::vector<size_t> config_done_enable_values(config_done_port.get_width(), 1);
+  fp << generate_verilog_constant_values(config_done_enable_values);
+  fp << ";" << std::endl;
+
+  fp << "\tend" << std::endl;
+  print_verilog_comment(fp, "----- End bitstream loading during configuration phase -----");
+}
+
+/********************************************************************
+ * Generate the stimuli for the top-level testbench 
+ * The simulation consists of two phases: configuration phase and operation phase
+ * Configuration bits are loaded serially. 
+ * This is actually what we do for a physical FPGA
+ *******************************************************************/
+static 
+void print_verilog_top_testbench_bitstream(std::fstream& fp,
+                                           const e_config_protocol_type& sram_orgz_type,
+                                           const BitstreamManager& bitstream_manager,
+                                           const std::vector<ConfigBitId>& fabric_bitstream) {
+  /* Branch on the type of configuration protocol */
+  switch (sram_orgz_type) {
+  case CONFIG_MEM_STANDALONE:
+    /* TODO */
+    break;
+  case CONFIG_MEM_SCAN_CHAIN:
+    print_verilog_top_testbench_configuration_chain_bitstream(fp, bitstream_manager, fabric_bitstream);
+    break;
+  case CONFIG_MEM_MEMORY_BANK:
+    /* TODO */
+    break;
+  default: 
+    VTR_LOGF_ERROR(__FILE__, __LINE__,
+                   "Invalid SRAM organization type!\n");
+    exit(1);
+  }
+}
+
+/********************************************************************
+ * The top-level function to generate a testbench, in order to verify: 
+ * 1. Configuration phase of the FPGA fabric, where the bitstream is 
+ *    loaded to the configuration protocol of the FPGA fabric
+ * 2. Operating phase of the FPGA fabric, where input stimuli are
+ *    fed to the I/Os of the FPGA fabric
+ *                                    +----------+
+ *                                    |   FPGA   |       +------------+
+ *                             +----->|  Fabric  |------>|            |
+ *                             |      |          |       |            |
+ *                             |      +----------+       |            |
+ *                             |                         | Output     | 
+ *   random_input_vectors -----+                         | Vector     |---->Functional correct?
+ *                             |                         | Comparator |    
+ *                             |      +-----------+      |            |
+ *                             |      |  Input    |      |            |
+ *                             +----->| Benchmark |----->|            |
+ *                                    +-----------+      +------------+
+ *
+ *******************************************************************/
+void print_verilog_top_testbench(const ModuleManager& module_manager,
+                                 const BitstreamManager& bitstream_manager,
+                                 const std::vector<ConfigBitId>& fabric_bitstream,
+                                 const e_config_protocol_type& sram_orgz_type,
+                                 const CircuitLibrary& circuit_lib,
+                                 const std::vector<CircuitPortId>& global_ports,
+                                 const AtomContext& atom_ctx,
+                                 const PlacementContext& place_ctx, 
+                                 const IoLocationMap& io_location_map,
+                                 const VprNetlistAnnotation& netlist_annotation,
+                                 const std::string& circuit_name,
+                                 const std::string& verilog_fname,
+                                 const std::string& verilog_dir,
+                                 const SimulationSetting& simulation_parameters) {
+
+  std::string timer_message = std::string("Write autocheck testbench for FPGA top-level Verilog netlist for '") + circuit_name + std::string("'");
+
+  /* Start time count */
+  vtr::ScopedStartFinishTimer timer(timer_message);
+
+  /* Create the file stream */
+  std::fstream fp;
+  fp.open(verilog_fname, std::fstream::out | std::fstream::trunc);
+
+  /* Validate the file stream */
+  check_file_stream(verilog_fname.c_str(), fp);
+
+  /* Generate a brief description on the Verilog file*/
+  std::string title = std::string("FPGA Verilog Testbench for Top-level netlist of Design: ") + circuit_name;
+  print_verilog_file_header(fp, title); 
+
+  /* Print preprocessing flags and external netlists */
+  print_verilog_include_defines_preproc_file(fp, verilog_dir);
+
+  /* Find the top_module */
+  ModuleId top_module = module_manager.find_module(generate_fpga_top_module_name());
+  VTR_ASSERT(true == module_manager.valid_module_id(top_module));
+
+  /* Preparation: find all the clock ports */
+  std::vector<std::string> clock_port_names = find_atom_netlist_clock_port_names(atom_ctx.nlist, netlist_annotation);
+
+  /* Start of testbench */
+  print_verilog_top_testbench_ports(fp, module_manager, top_module, 
+                                    atom_ctx, netlist_annotation, clock_port_names,
+                                    sram_orgz_type, circuit_name);
+
+  /* Find the clock period */
+  float prog_clock_period = (1./simulation_parameters.programming_clock_frequency());
+  float op_clock_period = (1./simulation_parameters.operating_clock_frequency());
+  /* Estimate the number of configuration clock cycles 
+   * by traversing the linked-list and count the number of SRAM=1 or BL=1&WL=1 in it.
+   * We plus 1 additional config clock cycle here because we need to reset everything during the first clock cycle
+   */
+  size_t num_config_clock_cycles = 1 + fabric_bitstream.size();
+
+  /* Generate stimuli for general control signals */
+  print_verilog_top_testbench_generic_stimulus(fp,
+                                               num_config_clock_cycles,
+                                               prog_clock_period,
+                                               op_clock_period,
+                                               VERILOG_SIM_TIMESCALE);
+
+  /* Generate stimuli for global ports or connect them to existed signals */
+  print_verilog_top_testbench_global_ports_stimuli(fp,
+                                                   module_manager, top_module,
+                                                   circuit_lib, global_ports);
+
+  /* Instanciate FPGA top-level module */
+  print_verilog_testbench_fpga_instance(fp, module_manager, top_module, 
+                                        std::string(TOP_TESTBENCH_FPGA_INSTANCE_NAME)); 
+  
+  /* Connect I/Os to benchmark I/Os or constant driver */
+  print_verilog_testbench_connect_fpga_ios(fp, module_manager, top_module,
+                                           atom_ctx, place_ctx, io_location_map,
+                                           netlist_annotation, 
+                                           std::string(), 
+                                           std::string(TOP_TESTBENCH_FPGA_OUTPUT_POSTFIX), 
+                                           (size_t)VERILOG_DEFAULT_SIGNAL_INIT_VALUE);
+
+  /* Instanciate input benchmark */
+  print_verilog_top_testbench_benchmark_instance(fp, 
+                                                 circuit_name,
+                                                 atom_ctx,
+                                                 netlist_annotation);
+
+  /* Print tasks used for loading bitstreams */
+  print_verilog_top_testbench_load_bitstream_task(fp, sram_orgz_type);
+
+  /* load bitstream to FPGA fabric in a configuration phase */
+  print_verilog_top_testbench_bitstream(fp, sram_orgz_type,
+                                        bitstream_manager, fabric_bitstream);
+
+  /* Add stimuli for reset, set, clock and iopad signals */
+  print_verilog_testbench_random_stimuli(fp, atom_ctx, 
+                                         netlist_annotation, 
+                                         clock_port_names, 
+                                         std::string(TOP_TESTBENCH_CHECKFLAG_PORT_POSTFIX),
+                                         BasicPort(std::string(TOP_TB_OP_CLOCK_PORT_NAME), 1));
+
+  /* Add output autocheck */
+  print_verilog_testbench_check(fp, 
+                                std::string(AUTOCHECKED_SIMULATION_FLAG),
+                                std::string(TOP_TESTBENCH_SIM_START_PORT_NAME),
+                                std::string(TOP_TESTBENCH_REFERENCE_OUTPUT_POSTFIX),
+                                std::string(TOP_TESTBENCH_FPGA_OUTPUT_POSTFIX),
+                                std::string(TOP_TESTBENCH_CHECKFLAG_PORT_POSTFIX),
+                                std::string(TOP_TESTBENCH_ERROR_COUNTER),
+                                atom_ctx,
+                                netlist_annotation, 
+                                clock_port_names,
+                                std::string(TOP_TB_OP_CLOCK_PORT_NAME));
+
+  /* Find simulation time */
+  float simulation_time = find_simulation_time_period(VERILOG_SIM_TIMESCALE,
+                                                      num_config_clock_cycles,
+										              1./simulation_parameters.programming_clock_frequency(),
+                                                      simulation_parameters.num_clock_cycles(),
+                                                      1./simulation_parameters.operating_clock_frequency());
+
+
+  /* Add Icarus requirement */
+  print_verilog_timeout_and_vcd(fp, 
+                                std::string(ICARUS_SIMULATOR_FLAG),
+                                std::string(circuit_name + std::string(AUTOCHECK_TOP_TESTBENCH_VERILOG_MODULE_POSTFIX)),
+                                std::string(circuit_name + std::string("_formal.vcd")), 
+                                std::string(TOP_TESTBENCH_SIM_START_PORT_NAME),
+                                std::string(TOP_TESTBENCH_ERROR_COUNTER),
+                                (int)simulation_time);
+
+
+  /* Testbench ends*/
+  print_verilog_module_end(fp, std::string(circuit_name) + std::string(AUTOCHECK_TOP_TESTBENCH_VERILOG_MODULE_POSTFIX));
+
+  /* Close the file stream */
+  fp.close();
+}
+
+} /* end namespace openfpga */

openfpga/src/fpga_verilog/verilog_top_testbench.h

@@ -0,0 +1,41 @@
+#ifndef VERILOG_TOP_TESTBENCH
+#define VERILOG_TOP_TESTBENCH
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+#include <string>
+#include <vector>
+#include "module_manager.h"
+#include "bitstream_manager.h"
+#include "circuit_library.h"
+#include "vpr_context.h"
+#include "io_location_map.h"
+#include "vpr_netlist_annotation.h"
+#include "simulation_setting.h"
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+void print_verilog_top_testbench(const ModuleManager& module_manager,
+                                 const BitstreamManager& bitstream_manager,
+                                 const std::vector<ConfigBitId>& fabric_bitstream,
+                                 const e_config_protocol_type& sram_orgz_type,
+                                 const CircuitLibrary& circuit_lib,
+                                 const std::vector<CircuitPortId>& global_ports,
+                                 const AtomContext& atom_ctx,
+                                 const PlacementContext& place_ctx, 
+                                 const IoLocationMap& io_location_map,
+                                 const VprNetlistAnnotation& netlist_annotation, 
+                                 const std::string& circuit_name,
+                                 const std::string& verilog_fname,
+                                 const std::string& verilog_dir,
+                                 const SimulationSetting& simulation_parameters);
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/fpga_verilog/verilog_writer_utils.h

@@ -14,6 +14,7 @@
 #include <string>
 #include "openfpga_port.h"
 #include "verilog_port_types.h"
+#include "circuit_library.h"
 #include "module_manager.h"
 
 /********************************************************************

openfpga/src/main.cpp

@@ -14,6 +14,7 @@
 #include "openfpga_setup_command.h"
 #include "openfpga_verilog_command.h"
 #include "openfpga_bitstream_command.h"
+#include "openfpga_sdc_command.h"
 #include "basic_command.h"
 
 #include "openfpga_title.h"
@@ -60,6 +61,9 @@ int main(int argc, char** argv) {
   /* Add openfpga bitstream commands */
   openfpga::add_openfpga_bitstream_commands(shell);
 
+  /* Add openfpga sdc commands */
+  openfpga::add_openfpga_sdc_commands(shell);
+
   /* Add basic commands: exit, help, etc. 
    * Note:
    * This MUST be the last command group to be added! 

openfpga/src/utils/openfpga_atom_netlist_utils.cpp

@@ -0,0 +1,41 @@
+/***************************************************************************************
+ * This file includes most utilized functions that are used to acquire data from 
+ * VPR atom netlist (users' netlist to implement)
+ ***************************************************************************************/
+
+/* Headers from vtrutil library */
+#include "vtr_log.h"
+#include "vtr_assert.h"
+#include "vtr_time.h"
+
+/* Headers from vtrutil library */
+#include "atom_netlist_utils.h"
+
+#include "openfpga_atom_netlist_utils.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+/***************************************************************************************
+ * Find the names of all the atom blocks that drive clock nets 
+ * This function will find if the block has been renamed due to contain sensitive characters
+ * that violates the Verilog syntax
+ ***************************************************************************************/
+std::vector<std::string> find_atom_netlist_clock_port_names(const AtomNetlist& atom_nlist,
+                                                            const VprNetlistAnnotation& netlist_annotation) {
+  std::vector<std::string> clock_names;
+
+  std::set<AtomPinId> clock_pins = find_netlist_logical_clock_drivers(atom_nlist);
+  for (const AtomPinId& clock_pin : clock_pins) {
+    const AtomBlockId& atom_blk = atom_nlist.port_block(atom_nlist.pin_port(clock_pin));
+    std::string block_name = atom_nlist.block_name(atom_blk);
+    if (true == netlist_annotation.is_block_renamed(atom_blk)) {
+      block_name = netlist_annotation.block_name(atom_blk);
+    }
+    clock_names.push_back(block_name);
+  }
+
+  return clock_names;
+}
+
+} /* end namespace openfpga */

openfpga/src/utils/openfpga_atom_netlist_utils.h

@@ -0,0 +1,24 @@
+#ifndef OPENFPGA_ATOM_NETLIST_UTILS_H
+#define OPENFPGA_ATOM_NETLIST_UTILS_H
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+#include <vector>
+#include <string>
+#include "atom_netlist.h"
+#include "vpr_netlist_annotation.h"
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+std::vector<std::string> find_atom_netlist_clock_port_names(const AtomNetlist& atom_nlist,
+                                                            const VprNetlistAnnotation& netlist_annotation);
+
+} /* end namespace openfpga */
+
+#endif

openfpga/src/utils/simulation_utils.cpp

@@ -0,0 +1,47 @@
+/********************************************************************
+ * This file include most utilized functions in generating simulations
+ * Note: function placed here MUST be generic enough for both SPICE
+ * and Verilog simulations!
+ *******************************************************************/
+#include <cmath>
+
+#include "simulation_utils.h"
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+
+/********************************************************************
+ * Compute the time period for the simulation
+ *******************************************************************/
+int find_operating_phase_simulation_time(const int& factor,
+                                         const int& num_op_clock_cycles,
+                                         const float& op_clock_period,
+                                         const float& timescale) {
+  /* Take into account the prog_reset and reset cycles 
+   * 1e9 is to change the unit to ns rather than second 
+   */
+  return (factor * num_op_clock_cycles * op_clock_period) / timescale; 
+}
+
+/********************************************************************
+ * Find the the full time period of a simulation, including
+ * both the programming time and operating time
+ * This is a generic function that can be used to generate simulation
+ * time period for SPICE/Verilog simulators
+ *******************************************************************/
+float find_simulation_time_period(const float &time_unit,
+                                  const int &num_prog_clock_cycles,
+                                  const float &prog_clock_period,
+                                  const int &num_op_clock_cycles,
+                                  const float &op_clock_period) {
+  float total_time_period = 0.;
+
+  /* Take into account the prog_reset and reset cycles */
+  total_time_period = (num_prog_clock_cycles + 2) * prog_clock_period + num_op_clock_cycles * op_clock_period;
+  total_time_period = total_time_period / time_unit;
+
+  return total_time_period;
+}
+
+} /* end namespace openfpga */

openfpga/src/utils/simulation_utils.h

@@ -0,0 +1,28 @@
+#ifndef SIMULATION_UTILS_H
+#define SIMULATION_UTILS_H
+
+/********************************************************************
+ * Include header files that are required by function declaration
+ *******************************************************************/
+
+/********************************************************************
+ * Function declaration
+ *******************************************************************/
+
+/* begin namespace openfpga */
+namespace openfpga {
+
+int find_operating_phase_simulation_time(const int& factor,
+                                         const int& num_op_clock_cycles,
+                                         const float& op_clock_period,
+                                         const float& timescale);
+
+float find_simulation_time_period(const float& time_unit,
+                                  const int& num_prog_clock_cycles,
+                                  const float& prog_clock_period,
+                                  const int& num_op_clock_cycles,
+                                  const float& op_clock_period);
+
+} /* end namespace openfpga */
+
+#endif

openfpga/test_blif/and.blif

@@ -0,0 +1,8 @@
+.model top
+.inputs a b
+.outputs c
+
+.names a b c
+11 1
+
+.end

openfpga/test_blif/and.v

@@ -0,0 +1,14 @@
+`timescale 1ns / 1ps
+
+module top(
+  a,
+  b,
+  c);
+
+input wire a;
+input wire b;
+output wire c;
+
+assign c = a & b;
+
+endmodule

openfpga/test_blif/s298.blif

@@ -1,90 +0,0 @@
-# Benchmark "s298" written by ABC on Tue Mar 12 09:40:31 2019
-.model s298
-.inputs clock G0 G1 G2
-.outputs G117 G132 G66 G118 G133 G67
-
-.latch        n21        G10 re      clock  0
-.latch        n26        G11 re      clock  0
-.latch        n31        G12 re      clock  0
-.latch        n36        G13 re      clock  0
-.latch        n41        G14 re      clock  0
-.latch        n46        G15 re      clock  0
-.latch        n51        G66 re      clock  0
-.latch        n55        G67 re      clock  0
-.latch        n59       G117 re      clock  0
-.latch        n63       G118 re      clock  0
-.latch        n67       G132 re      clock  0
-.latch        n71       G133 re      clock  0
-.latch        n75        G22 re      clock  0
-.latch        n80        G23 re      clock  0
-
-.names n56 n57 G10 n63
-0-0 1
-11- 1
-.names G15 G11 G13 G22 G14 G12 n56
-01---- 1
-0-0--- 1
-0--0-- 1
-0---1- 1
-0----1 1
--11000 1
-.names G14 G13 G12 G118 G11 n57
-01--- 1
-100-0 1
-1-11- 1
--1-1- 1
-.names n56 n59_1 G10 n67
-0-0 1
-11- 1
-.names G14 G13 G12 G132 G11 n59_1
-100-0 1
-11-1- 1
-1-11- 1
-.names G0 G10 n21
-00 1
-.names G10 G11 G0 G12 G13 n26
-010-- 1
-1001- 1
-100-0 1
-.names G12 G0 G11 G10 n31
-0011 1
-100- 1
-10-0 1
-.names G13 G0 G11 G12 G10 n36
-00111 1
-1001- 1
-1010- 1
-10--0 1
-.names n65 G14 G0 n41
-000 1
-110 1
-.names G23 G10 G13 G11 G12 n65
-1---- 0
--1100 0
-.names G0 n56 n46
-00 1
-.names n56 G66 G14 G13 G12 n51
-111-1 1
-11-1- 1
-1-01- 1
-.names n56 G13 G14 G11 G67 G12 n55
-1000-- 1
-10-1-0 1
-111-1- 1
-1-1-11 1
-.names n56 G13 G117 G14 G12 G11 n59
-10-0-- 1
-10--01 1
-1111-- 1
-1-111- 1
-.names n56 G14 G12 G13 G133 G11 n71
-1010-1 1
-111-1- 1
-11-11- 1
-.names G2 G22 G0 n75
-010 1
-100 1
-.names G1 G23 G0 n80
-010 1
-100 1
-.end

openfpga/test_openfpga_arch/k6_frac_N10_40nm_openfpga.xml

@@ -168,11 +168,11 @@
        <input_buffer exist="true" circuit_model_name="INVTX1"/>
        <output_buffer exist="true" circuit_model_name="INVTX1"/>
        <pass_gate_logic circuit_model_name="TGATE"/>
-       <port type="input" prefix="pReset" size="1" is_global="true" default_val="0" is_reset="true" is_prog="true"/>
+       <port type="input" prefix="pReset" lib_name="reset" size="1" is_global="true" default_val="0" is_reset="true" is_prog="true"/>
        <port type="input" prefix="D" size="1"/>
        <port type="output" prefix="Q" size="1"/>
        <port type="output" prefix="Qb" size="1"/>
-       <port type="clock" prefix="prog_clk" size="1" is_global="true" default_val="0" is_prog="true"/>
+       <port type="clock" prefix="prog_clk" lib_name="clk" size="1" is_global="true" default_val="0" is_prog="true"/>
     </circuit_model>
     <circuit_model type="iopad" name="iopad" prefix="iopad" spice_netlist="${OPENFPGA_PATH}/openfpga_flow/SpiceNetlists/io.sp" verilog_netlist="${OPENFPGA_PATH}/openfpga_flow/VerilogNetlists/io.v">
       <design_technology type="cmos"/>
@@ -232,6 +232,7 @@
 </openfpga_architecture>
 <openfpga_simulation_setting>
   <clock_setting>
+    <!--operating frequency="auto" num_cycles="auto" slack="0.2"/-->
     <operating frequency="200e6" num_cycles="auto" slack="0.2"/>
     <programming frequency="10e6"/>
   </clock_setting>

openfpga/test_script/and_k6_frac.openfpga

@@ -0,0 +1,53 @@
+# Run VPR for the s298 design
+vpr ./test_vpr_arch/k6_frac_N10_40nm.xml ./test_blif/and.blif --clock_modeling route #--write_rr_graph example_rr_graph.xml
+
+# Read OpenFPGA architecture definition
+read_openfpga_arch -f ./test_openfpga_arch/k6_frac_N10_40nm_openfpga.xml
+
+# Write out the architecture XML as a proof
+#write_openfpga_arch -f ./arch_echo.xml
+
+# Annotate the OpenFPGA architecture to VPR data base
+link_openfpga_arch --activity_file ./test_blif/and.act #--verbose
+
+# Check and correct any naming conflicts in the BLIF netlist
+check_netlist_naming_conflict --fix --report ./netlist_renaming.xml
+
+# Apply fix-up to clustering nets based on routing results
+pb_pin_fixup --verbose
+
+# Apply fix-up to Look-Up Table truth tables based on packing results
+lut_truth_table_fixup #--verbose
+
+# Build the module graph 
+#  - Enabled compression on routing architecture modules
+#  - Enable pin duplication on grid modules 
+build_fabric --compress_routing --duplicate_grid_pin #--verbose
+
+# Repack the netlist to physical pbs
+# This must be done before bitstream generator and testbench generation
+# Strongly recommend it is done after all the fix-up have been applied
+repack #--verbose
+
+# Build the bitstream 
+#  - Output the fabric-independent bitstream to a file
+fpga_bitstream --verbose --file /var/tmp/xtang/openfpga_test_src/fabric_indepenent_bitstream.xml
+
+# Write the Verilog netlist for FPGA fabric
+#  - Enable the use of explicit port mapping in Verilog netlist
+write_fabric_verilog --file /var/tmp/xtang/openfpga_test_src/SRC --explicit_port_mapping --include_timing --include_signal_init --support_icarus_simulator --print_user_defined_template --verbose
+
+# Write the Verilog testbench for FPGA fabric
+#  - We suggest the use of same output directory as fabric Verilog netlists
+#  - Must specify the reference benchmark file if you want to output any testbenches
+#  - Enable top-level testbench which is a full verification including programming circuit and core logic of FPGA
+#  - Enable pre-configured top-level testbench which is a fast verification skipping programming phase
+#  - Simulation ini file is optional and is needed only when you need to interface different HDL simulators using openfpga flow-run scripts
+write_verilog_testbench --file /var/tmp/xtang/openfpga_test_src/SRC --reference_benchmark_file_path /var/tmp/xtang/and.v --print_top_testbench --print_preconfig_top_testbench --print_simulation_ini /var/tmp/xtang/openfpga_test_src/simulation_deck.ini
+
+# Write the SDC files for PnR backend
+#  - Turn on every options here 
+write_pnr_sdc --file /var/tmp/xtang/openfpga_test_src/SDC 
+
+# Finish and exit OpenFPGA
+exit

openfpga/test_script/s298.openfpga

@@ -1,14 +0,0 @@
-# Run VPR for the s298 design
-vpr ./test_vpr_arch/k6_N10_40nm.xml ./test_blif/s298.blif
-
-# Read OpenFPGA architecture definition
-read_openfpga_arch -f ./test_openfpga_arch/k6_N10_40nm_openfpga.xml
-
-# Annotate the OpenFPGA architecture to VPR data base
-link_openfpga_arch
-
-# Check and correct any naming conflicts in the BLIF netlist
-check_netlist_naming_conflict --fix --report ./netlist_renaming.xml
-
-# Finish and exit OpenFPGA
-exit

openfpga/test_script/s298_k6_frac.openfpga

@@ -1,41 +0,0 @@
-# Run VPR for the s298 design
-vpr ./test_vpr_arch/k6_frac_N10_40nm.xml ./test_blif/s298.blif --write_rr_graph example_rr_graph.xml
-
-# Read OpenFPGA architecture definition
-read_openfpga_arch -f ./test_openfpga_arch/k6_frac_N10_40nm_openfpga.xml
-
-# Write out the architecture XML as a proof
-#write_openfpga_arch -f ./arch_echo.xml
-
-# Annotate the OpenFPGA architecture to VPR data base
-link_openfpga_arch #--verbose
-
-# Check and correct any naming conflicts in the BLIF netlist
-check_netlist_naming_conflict --fix --report ./netlist_renaming.xml
-
-# Apply fix-up to clustering nets based on routing results
-pb_pin_fixup --verbose
-
-# Apply fix-up to Look-Up Table truth tables based on packing results
-lut_truth_table_fixup #--verbose
-
-# Build the module graph 
-#  - Enabled compression on routing architecture modules
-#  - Enable pin duplication on grid modules 
-build_fabric --compress_routing --duplicate_grid_pin #--verbose
-
-# Repack the netlist to physical pbs
-# This must be done before bitstream generator and testbench generation
-# Strongly recommend it is done after all the fix-up have been applied
-repack --verbose
-
-# Build the bitstream 
-#  - Output the fabric-independent bitstream to a file
-fpga_bitstream --verbose --file /var/tmp/xtang/openfpga_test_src/fabric_indepenent_bitstream.xml
-
-# Write the Verilog netlit for FPGA fabric
-#  - Enable the use of explicit port mapping in Verilog netlist
-write_fabric_verilog --file /var/tmp/xtang/openfpga_test_src --explicit_port_mapping --include_timing --include_signal_init --support_icarus_simulator --print_user_defined_template --verbose
-
-# Finish and exit OpenFPGA
-exit

openfpga/test_vpr_arch/k6_frac_N10_40nm.xml

@@ -45,19 +45,22 @@
     </model>
   </models>
   <tiles>
+    <!-- Do NOT add clock pins to I/O here!!! VPR does not build clock network in the way that OpenFPGA can support
+         If you need to register the I/O, define clocks in the circuit models
+         These clocks can be handled in back-end
+     -->
     <tile name="io" capacity="8" area="0">
       <equivalent_sites>
         <site pb_type="io"/>
       </equivalent_sites>
       <input name="outpad" num_pins="1"/>
       <output name="inpad" num_pins="1"/>
-      <clock name="clock" num_pins="1"/>
       <fc in_type="frac" in_val="0.15" out_type="frac" out_val="0.10"/>
       <pinlocations pattern="custom">
-        <loc side="left">io.outpad io.inpad io.clock</loc>
+        <loc side="left">io.outpad io.inpad</loc>
-        <loc side="top">io.outpad io.inpad io.clock</loc>
+        <loc side="top">io.outpad io.inpad</loc>
-        <loc side="right">io.outpad io.inpad io.clock</loc>
+        <loc side="right">io.outpad io.inpad</loc>
-        <loc side="bottom">io.outpad io.inpad io.clock</loc>
+        <loc side="bottom">io.outpad io.inpad</loc>
       </pinlocations>
     </tile>
     <tile name="clb" area="53894">
@@ -145,7 +148,10 @@
     <pb_type name="io">
       <input name="outpad" num_pins="1"/>
       <output name="inpad" num_pins="1"/>
-      <clock name="clock" num_pins="1"/>
+      <!-- Do NOT add clock pins to I/O here!!! VPR does not build clock network in the way that OpenFPGA can support
+           If you need to register the I/O, define clocks in the circuit models
+           These clocks can be handled in back-end
+       -->
       <!-- A mode denotes the physical implementation of an I/O 
            This mode will be not packable but is mainly used for fabric verilog generation   
         -->

vpr7_x2p/vpr/SRC/fpga_x2p/verilog/verilog_testbench_utils.cpp

@@ -479,7 +479,6 @@ void print_verilog_testbench_shared_ports(std::fstream& fp,
   /* Validate the file stream */
   check_file_handler(fp);
 
-
   /* Instantiate register for inputs stimulis */
   print_verilog_comment(fp, std::string("----- Shared inputs -------"));
   for (const t_logical_block& lb : L_logical_blocks) {